Traveller 5E

[swordtart]

No, a roll of 2d6 has 12 outcomes, a roll of "d66" has 36 outcomes, each with a flat percentage of ~3%

Not the same thing at all, stop for a moment and think about what you are saying.

We have different definitions of outcome.

An 8+ target number has a 42% chance of success with no modifiers, if you want to consider it as two potential outcomes then success is 42%, fail is 58%.

Exceptional is usually for more than 2 standard deviations, but the limits of the 2d6 roll don't allow for that.

By what criteria? Exceptional on p61 is either rolling 6 more or less than the target. With a default 8+ you can get that rolling snake eyes. Entirely possible on 2d6 without any modifiers. Once you count modifiers and effect of 6 one way or the other is entirely possible.

We are not talking about probability theory we are talking the rolls in Traveller (and specifically MGT2 Traveller in my case).

Six successful outcomes, each with their own percentage chance of occurrence.

Not sure what you mean, 3 of the outcomes on the Chain Table are failures and apply a negative modifier.

Boon and bane are a great mechanic, what isn't so great is a PC with a DM of+10 rolling against a standard target of 8... this is especially noticeable in space combat.

I agree space combat applies too many modifiers. Space combat is a bit pants in general though (and the outcome is a bit digital either the ship is destroyed and all characters die or it isn't and it just becomes spreadsheets in space managing the repairs) so I avoid it.

Normal combat requires you to make choices to get those ridiculous modifiers (aim for 2 turns gets you +6, but that means not firing for 2 turns).

I think that greater use of the boon/bane mechanic in place of stacking DMs would create a better resolution system for 2d6.

I don't allow too much unearned stacking. If the player is smart enough to engineer the situation so that they can bring multiple DMs to bear then that is intelligent play. It is generally easier to accommodate an unexpected success by a character than a fluffed roll when the plot was assuming success. However regardless, the dice decide we abide (unless we decide to deus ex machina).

Or use percentile dice.

The game assumes a standard difficulty of 8+, perhaps shifting the standard difficulty to 12+ would make for a "better game".

8+ is supposed to pose a moderate challenge to a "trained professional".

"Trained" is skill 0. I know there are others that apply skill inflation and contest that a professional should be skill level 2 or higher, but a professional is someone who does something for a living not someone who is a subject matter expert.

On p90-91 we are told a "Skilled Professional" has 2 or 3 levels in skills related to their profession. I took that to mean they have a total of 2-3 levels in those skills, not that they have 2-3 levels in EACH of those skills.

You can spend 4 years in basic training and at the end of it you are a trained professional (it's your job). You may end up with no skill above level 0. You are still a trained professional.

You don't get to tell me how to play or referee, all I am pointing out is what is allowed by the rules as written. Traveller characters with cyberware and other augments are becoming street level super heroes rather than blue collar space workers...

I didn't think I was telling you. I offered a suggested solution to a problem that you implied you were experiencing. Personally I am not convinced someone with KCr100's in cyberware is in any way a "blue-collar worker". Blue collar workers are usually the background characters in space opera. A prosthetic arm partially paid for by the company due to some industrial accident is blue collar. A cyber arm makes you "Molly Millions" and a gun for hire, not a salaryman.

The trope of cyberpunk is for the majority of main characters not to even have extensive cyber ware, the chromed ones are usually the target or the big bad.

You should see how augmented some of the characters in my Culture setting are.

"Your" setting.

A 5 term marine and a 1 term scout both face the same chances when their opponent snipes them from surprise with a laser rifle...

Exactly my point and entirely as it should be, the sniper has leveraged every advantage and should not really miss, it isn't hard to hit an unsuspecting target from even quite far away if you get to choose the time of attack. The marine however should be using his training to avoid being in situation where he is the target of a sniper. That is less about skill levels and more about players making dumb decisions. We can be charitable and allow them a Recon skill check to allow them to identify the likely sites of snipers (it's always a church tower in the movies but that isn't necessarily true in real-life) even if they don't know about a specific immediate threat. If they insist in irritating the local crime boss, establish a routine of sitting at a particular cafe at a particular time that is within 200m of an abandoned church then they are not behaving like a 5-term marine.

The problem as I see it is the game allowing easy access to very high DMs via equipment and augmentation.

It is not the game doing that, that is a referee decison.

Only Agent and Army careers allow 1 in 6 chance of a TL12 augmentation as a benefit, it is still for the referee to allow any specific augmentation or equipment. As the limit is KCr75 there are not many that are in budget. The chance of getting the benefit twice (where you can exceed that limit) is naturally lower (on average you would need to take 7 non-cash benefits to get it twice). Other careers offer an equipment benefit but it is often quite poor and the cash is often a better option. If you take it then getting a +1 or +2 sounds a reasonable compensation.

Personally I only allow equipment from the core rules as mustering out benefits. Stuff from CSC etc. needs to be bought in game or is an adventure reward (usually compensating for something the character cannot do rather than making them infallible in something they can already do pretty well). Very little equipment allows you to stack onto a skill bonus. Again I generally only allow 1 piece of equipment to be used at a time. I think DM+2 is the most that a toolkit can provide and that already includes an expert system.

If you are really lucky you might get a skill augment and a toolkit, but that just means that in a single specialism of a skill you are 'da man'. If you have Electronics(Computers)-2 get +1 for a stat and another +2 for equipment and +1 for an augment for a total DM+6, that is your USP, so go team you (you can still fail an average check 1 time in 36 which still makes you pretty fallible in the long term). The referee does not have to spoon feed you computer based adventures though.

"You spend 4 weeks refactoring the ISS data management system. Roll a skill check - ok you get effect x KCr1 as a bonus. You are leaving the building when you get a weird feeling - Roll an Average Recon check - ooh effect -3, never mind. You see a heavy set, rough looking type step out behind a support pillar. Ok, you turn around only to see another man directly behind you holding a stubby pistol. He shrugs sympathetically before your body is wracked in pain from the stun pistol. Ouch, it completely drops your endurance and you crumple in a twitching heap and the world goes dark."

Plenty of good fiction throws someone who is a solid genius in some skill into totally the wrong environment to see them flounder and eventually overcome. These are the good stories.

 

[WHULorigan]

You don't have 36 "outcomes" you have 12 "outcomes" with a not quite standard distribution spread. A roll of 1 and 3 is different to a roll of 3 and 1 yet the outcome is 4.

To have 36 outcomes you would have to read the dice as die a then die b in which case you get a flat probability of 36 outcomes rather then the flat probability of the d20.

The "bell curve" of 2d6 is not a bell curve, that said 67% of the rolls you make will be 5 to 9 (within the one standard deviation bound), two standard deviation give a spread of 2 to 12 so your "outcomes" on your die roll fall into
2 standard deviations below mean - 2->4
1 standard deviation below mean - 5->6
mean 7
1 standard deviation above mean - 8->9
2 standard deviations above mean - 10->12

By setting the standard task target number as 8 you get a flat percentage chance of succeeding on your 2d6 roll of 42%. DMs now have a massive effect on this, DMs of 1->3 are just about containable, but as Endie says it is now typical for a PC to have a minimum of a +5 bonus (equipment, expert program, augmentation) before adding characteristic and skill level.

Using a d20 you would set the target number as 13.

To map Traveller difficulty target numbers to the target numbers of a d20 roll we get something like this:

Traveller target number	2	4	6	8	10	12
d20 target number equivalence	1	2	6	13	17	20

Sigtrygg, might it be helpful to reevaluate the "2D6" target number spread considering a 1-15 range (since Stats can range that far)? In a certain sense, Traveller is actually Hexadecimal, not duodecimal. We only roll 2D6, but the expectation is that base stats, base stat bonuses inherently related to their standard distribution, and target numbers will range higher.

Allow the normal "smooth" curve to be 1-13. 14-15 breaks the curve on the high end. How does that affect the mechanics?
(I am talking in very general terms here.)


Note that most other 2D6 based systems (non-Traveller-related) have 12 as the maximum of their target, stat, and roll ranges.

 

[Endie]

2d10 with targets of 12 for standard tasks, and 16, 20, 24, 28 for hard to impossible ones. That means that the hardest tasks really feel exceptional, even for those stacking advantages. It also means that success in standard tasks is expected for the maxed-out player, but even the Super Bowl QB misses a few.

To those saying you just change the game by removing opinions instead: you’re just agreeing that DM inflation is out of hand: all we are doing is arguing over methodology. At least by widening the range of results you allow players to indulge in those sci-fi options the game provides.

 

[swordtart]

I am going to explain why I think there are more than 12 outcomes.

When rolling 2d6 in Traveller the identity of the two dice is irrelevant (if it is then it is a d66 roll). If one die is red and one die is blue it doesn't matter to the outcome if red rolls 3 and blue rolls 2 or vice versa. You still get a 2 and a 3. On this basis the only outcomes would be 1-1, (1-2 or 2-1), (1-3 or 3-1)... (6-4 or 4-6), (6-5 or 5-6), and 6-6. So this allows 21 unique outcomes.

We could go further and argue that the actual numbers are irrelevant and it is only the score that matters (1-5, 2-4, 3-3, 4-2 5-1) are all identical and equal 6. This gives us 11 possible outcomes.

The complication is that there are mechanics in the game like Boons and Banes that do depend on the values on specific dice rather than just the combined total. I am going to simplify the boon as instead of rolling 3 dice and choosing the best 2, I will instead re-roll the worst dice and keep the better score. Hopefully it will be clear that this is functionally identical as we have established the the identity of the dice is irrelevant*.

If my target is 8+ on the dice and I roll a 3 and a 4, I fail. If I have a boon I can re-roll the 3 and hope to get a 4+. If I start with a 6 and a 1 on the other hand I still fail, but I can reroll the 1 and only need a 2+ to make the roll.
With the same initial roll if I have a bane then I have to re-roll the 4, but I can still make may target if I get 5+. If I start with the 6 and 1 however I need to reroll the 6 and my chance of getting a 8+ is nil.

Both initial rolls had a total value of 7 but they are not the same outcome as they have different effects in the event a boon or bane gives a "re-roll".

Another example would be weapons with the "Deadly" advantage. For these a roll of 1-3 is counted as 3. If we have a deadly blade and again roll 7. With a 1 and 6 we count the 1 as 3 and the total would be 9 points of damage. If we rolled a 3 and 4 it would remain at 7 points. Both rolls were identical totals but they are not equivalent outcomes as the specific value on each dice might have caused the special Deadly effect to trigger.

Doubtless there are other wonders and artifacts that also monkey about with dice to make even a simple 2d6 more varied, flexible and interesting than might be initially supposed. It will be far less swingy than a D20 for which each result is equally probable.

*I do it this way in game as it reduces the number of dice clattering about, (inevitably one rolls on the floor) and it also creates a bit of extra tension as a near miss is turned into a success. The boon or bane flipping the outcome generally gets more story focus as well in this way and players can relate to it more when it saves (or dooms) them vs. just being another dice in the mix. If the effect is irrelevant and the initial roll is a success then no re-roll is needed and we save a little time.

 

[Sigtrygg]

I am going to explain why I think there are more than 12 outcomes.

When rolling 2d6 in Traveller the identity of the two dice is irrelevant (if it is then it is a d66 roll). If one die is red and one die is blue it doesn't matter to the outcome if red rolls 3 and blue rolls 2 or vice versa. You still get a 2 and a 3. On this basis the only outcomes would be 1-1, (1-2 or 2-1), (1-3 or 3-1)... (6-4 or 4-6), (6-5 or 5-6), and 6-6. So this allows 21 unique outcomes.

This is simply incorrect

Result	2	3	4	5	6	7	8	9	10	11	12
No. of possible ways	1	2	3	4	5	6	5	4	3	2	1

There are 36 outcomes

We could go further and argue that the actual numbers are irrelevant and it is only the score that matters (1-5, 2-4, 3-3, 4-2 5-1) are all identical and equal 6. This gives us 11 possible outcomes.

It is the number of possible ways to roll the final number that gives the probability of rolling that result

Result	2	3	4	5	6	7	8	9	10	11	12
probability	1/36	2/36	3/36	4/36	5/36	6/36	5/36	4/36	3/36	2/36	1/36

To get the probability of rolling a given result or higher sum the probabilities for that result and all above it.

So 8+ is (5+4+3+2+1)/36 = 15/36 which is... 42%

The complication is that there are mechanics in the game like Boons and Banes that do depend on the values on specific dice rather than just the combined total. I am going to simplify the boon as instead of rolling 3 dice and choosing the best 2, I will instead re-roll the worst dice and keep the better score. Hopefully it will be clear that this is functionally identical as we have established the the identity of the dice is irrelevant*.

So you are changing the mechanic. Your house rule method gives a different probability distribution than rolling 3d and keeping the highest 2/lowest 2. Further discussion is comparing you house rule to the rules as written.

*I do it this way in game as it reduces the number of dice clattering about, (inevitably one rolls on the floor) and it also creates a bit of extra tension as a near miss is turned into a success. The boon or bane flipping the outcome generally gets more story focus as well in this way and players can relate to it more when it saves (or dooms) them vs. just being another dice in the mix. If the effect is irrelevant and the initial roll is a success then no re-roll is needed and we save a little time.

You are changing the game mechanic and the probability range the game mechanic is based on.

 

[Talon Greyfeather]

The 2d6 system breaks down when you have a bonus of +6 before adding characteristic and skill for a standard target number of 8+

Then the issue is the referee allowing the characters to acquire the +6 bonus in the first place - just like it is in any other RPG.

 

[rinku]

I look at it this way. When there are a lot of positive modifiers, either from skill or from circumstance, yes. You can get to automatic success.

Why is that a problem? For me, it makes FAR more sense than old school d20 or d% giving a large change of failure for specialty skills by experienced characters. (Sneaking thieves, you know I'm talking to you).

The Effect rule is what saves things for me. There's your distinction between +6 with a low roll and +6 with a high roll.

Bang! Rolled a 2 to hit... no bonus damage. Bang! rolled a 12 to hit... +10 damage!

Profession (Carpenter) with a +6: rolled a 2, you MAKE the table, but it's a bit wobbly and the finish is meh; rolled a 12, it's a damn museum quality master piece, or so well made your great grandkids will be still putting their feet up on it and being yelled at for it by your grandkid.

 

[Endie]

Then the issue is the referee allowing the characters to acquire the +6 bonus in the first place - just like it is in any other RPG.

As I said: by saying this you agree that the game is broken, RAW. You just want to try to fix the rules by removing options that the rulebooks explicitly provide, thus removing player options. By changing the dice you allow players to face at least occasional challenge while also allowing the additional variety that the game provides.

The tutting approach of judging players for wanting to have fun, sci-fi options in a game based thousands of years in the future (or judging GMs for allowing them to explore such sci-fi scenarios) always bemuses me.

On the one side people talk about the ubiquity of the Traveller ruleset, and how it can be used to run a game in a huge variety of sci-fi styles. And on the other we have people saying “oh but don’t let your players have cyberware or advanced medical equipment you incompetent excuse for a GM!”

 

[NOLATrav]

I feel like people in this discussion are missing the forest for the trees.

Augments, mods, high TL equipment and all the rest don’t exist to create auto-successes. They exist so the PCs can do the occasional Impossible Thing. That’s why there’s a scale of target numbers up to Impossible. If you let your PCs use that stuff when they don’t really need it, well yeah, they will always succeed. At that point you shouldn’t even ask for a roll. And that’s probably a boring story.

2D6 isn’t really broken but it is limited. GURPS did 3d6 but 4D6 with Average TN=16 is interesting… you’re really wanting that +1 or +2 DM when you’re lacking decent gear but then Formidable 20+ and Impossible 22+ are exactly what they say without the proper gear for the job. And that’s probably a very exciting story.

Refs - use the Difficulty scale! Not everything is nor should be 8+ !!!

 

[Endie]

I feel like people in this discussion are missing the forest for the trees.

Augments, mods, high TL equipment and all the rest don’t exist to create auto-successes. They exist so the PCs can do the occasional Impossible Thing. That’s why there’s a scale of target numbers up to Impossible. If you let your PCs use that stuff when they don’t really need it, well yeah, they will always succeed. At that point you shouldn’t even ask for a roll. And that’s probably a boring story.

2D6 isn’t really broken but it is limited. GURPS did 3d6 but 4D6 with Average TN=16 is interesting… you’re really wanting that +1 or +2 DM when you’re lacking decent gear but then Formidable 20+ and Impossible 22+ are exactly what they say without the proper gear for the job. And that’s probably a very exciting story.

Refs - use the Difficulty scale! Not everything is nor should be 8+ !!!

Sure, but the players will get slightly cynical after a time when every task turns out to be at Impossible. Without at least the chance of mishap, we change the game fundamentally. At that point, we are playing a narrative, almost diceless game, where the players succeed and the only question is how successful they are.

You dismiss those who don’t agree with you as not seeing the wood for the trees. I would suggest that, in fact, every single person so far in this thread has agreed (at least by implication) that Traveller 2D6 quickly breaks when played as written in the rulebooks. Changing the dice is the simplest fix, but those wishing to remove potential DMs by excising large portions of the Core rulebook, CSC and more are right that that will work, albeit with a lot more haggling with players.

All those blaming referees and players for wanting to use the equipment and cyberware provided in the rules are saying, effectively, “2D6 can’t handle the game as written.” I accept that there are people emotionally committed to the very concept of 2D6 who will never accept that reintroducing chance while maintaining scope is desirable: it’s just my own opinion, and I wholly accept that in matters of gaming and fun, de gustibus non est disputandum!

 

[swordtart]

This is simply incorrect
<snip>

No-one was arguing the probabilities. Anyone who has ever played an RPG game using dice probably knows this (and likely spent hours in their room generating probability analyses to min-max their character or WH40K squad). I was taught it in secondary school. However as you kept insisting there were 12 possible "outcomes" whereas even by your own argument there are only 11 led me to assume we were misunderstanding each other.

There seems to be confusion between outcomes and events. Outcomes in probability are all the possible combinations of dice results. There are 36 outcomes just as there are for a d66 each with exactly the same probability (1/36). What you decide to do with the numbers in your experiment does not affect the probabilities of those numbers occurring. Events are the number of results of interest from your outcomes. There are 11 events if you add the dice together and group them by score each event may have a different probability (as is the case with 2d6).

So you are changing the mechanic. Your house rule method gives a different probability distribution than rolling 3d and keeping the highest 2/lowest 2. Further discussion is comparing you house rule to the rules as written.

You are changing the game mechanic and the probability range the game mechanic is based on.

I was hoping you would be able to see that I was not actually changing the game mechanic, simply restructuring how it was set out to simplify the argument. There is no actual difference in probabilities of any specific result between throwing three dice and discarding one of them vs rolling two dice and re-rolling one of the two and choosing the result from that dice you prefer. Both use 3 dice rolls, both end up choosing the best (or worse) two of the three. If I had said you had to discard one dice before re-rolling it and take the new result then the mechanics would not be equivalent.

If you cannot see that then the conversation is indeed pointless.

Perhaps you will be able to accept that a 2d6 roll in traveller is sometimes practically a 3d6 roll and even by your argument has more that 11 outcomes even though only 11 different numbers can be generated. Further due to other mechanics in the game (like Deadly) not all D6 rolls are equal. In some cases a roll of snake eyes is always a fail regardless of what number you finally end up with. A flat statistical model of the sum of 2 dice shorn of context does not describe that complexity.

Remember this discussion had largely devolved into "2d6 Bad d20 Good". I was merely trying to point out that the 2d6 system is more than a badly structured percentile system and is capable of nuance despite the supposed limitation of only being able to generate 11 numbers on 2d6. In reality MGT2 is not a pure 2d6 system it is an d6 based system (as it doesn't use polyhedral dice), but sometimes you will be rolling 2 dice for a skill, sometimes 3 dice and sometimes the values on those dice will change for specific reasons and you will probably be adding to the rolled value. 2d6 system is a heritage term.

For d20 a +1 DM always moves the probability of achieving a particular total by 5% regardless of that target number (notwithstanding the boundary conditions of 1 and 20).
For the 2d6 system a +1 DM can changes the probability by a different amount depending on what your target number was.

5th edition also has an advantage/disadvantage mechanic. It doesn't make the D20 any less swingy though so it's effects can be . It also allows adding other types of dice to the d20 for specific effects. This provides more variety but combinations of DMs can be just as impactful as adding adding +6 to a 2d6 roll. If you are a reasonable level character (equivalent to a Traveller 5-term) using a magic sword+2 with +2 or a high stat, and expertise plus a few extras for bardic inspiration and bless (equivalent to special equipment) you could easily be adding +15 to your skill roll. We don't know what 5th Edition Traveller will be doing, but if it is not using equivalent mechanics then it isn't really a 5th edition port.

2d6 reinforces (but does not overwhelm) the average, but it does mean that if you too generous with your positive DMs, not diligent in enforcing the conditions required for those DM's to apply or not applying appropriate negative DMs your game can "run hot". Players that want to "win" will focus on their characters strengths and avoid situations where they are weak. There are fewer worthwhile stories there as there is less challenge and it is for the referee to put them outside their comfort zone (while allowing them to shine when they are the right person for the job).

Traveller (and contemporary games like 1st edition D&D) used to be about dealing with the challenge presented. This started at character generation as you got very little choice and even those choices were subject to random fluctuation. Referees were expected to be impartial but fair. Survival was a challenge and when you succeeded in a mission you felt good. When you failed, you sucked it up and endured the consequences ("Dingo is too wounded to continue, we'll have to leave him here and go on alone").
Modern sensibilities seem to be focussed on what challenges the players would like to face, the referee is expected to support their choices. Success is expected and failure is shorn of real consequence ("We'll have long rest and get all our HP and abilities back"). That changes the nature of the game far more than what dice mechanic you happen to use.

 

[swordtart]

As I said: by saying this you agree that the game is broken, RAW. You just want to try to fix the rules by removing options that the rulebooks explicitly provide, thus removing player options. By changing the dice you allow players to face at least occasional challenge while also allowing the additional variety that the game provides.

Every item in the game is optional. If you apply all optional positives and ignore all optional negatives you will get a different game. Not a better game or a worse game just a different one. No one is judging you for wanting to do that. If you do that and it breaks the game it is not however an indication the game is broken. Rule zero applies.

The tutting approach of judging players for wanting to have fun, sci-fi options in a game based thousands of years in the future (or judging GMs for allowing them to explore such sci-fi scenarios) always bemuses me.

If you and your players want to have everyone with +6 to skills that is fine. It is however a little implausible to have characters who can do a routine thing today unable to do it tomorrow under the exact same circumstances because of "reasons".

I am saying if you want your players to be Dan Dare pilot of the future who can hot wire a Venusian Cruiser with a paperclip, it seems a little implausible (and pointless) that Dan might crash land his own shuttle unless the local circumstances are unfavourable enough to disadvantage him (he is rushed, there is bad weather, Jimbo installed the avionics incorrectly etc.).

We make rolls when the results are interesting. Forcing players to make rolls for routine thing that might randomly derail the story is not interesting to every table. Some tables like chaos and they may wish to check even routine activity just in case it throws up a hilarious incident where "Kryten slips on a bar of soap". Others will want to focus on their characters impacting the universe but may wish to throw in the odd quirky session "That day Dave went to the fish market" to keep things fresh.

On the one side people talk about the ubiquity of the Traveller ruleset, and how it can be used to run a game in a huge variety of sci-fi styles. And on the other we have people saying “oh but don’t let your players have cyberware or advanced medical equipment you incompetent excuse for a GM!”

Currently all this is already possible in the RAW. Your argument seems to be that we should change the rules because they break your game.

I had a character that died in the first 10 minutes of our first session of Aftermath when getting out of a bunk bed. The referee wanting to show case the skill system and made me roll to see if I fell (falling while climbing was in the rules). I slipped, fell 6 feet, rolled on a table and critically failed braking my neck. It was memorable (and hilarious at the time) and we talked about it for weeks afterwards, but replacing my character proved to be an issue within the structure of that nights session (it took about an hour to roll up a character and I would be out of the game all that time). Due to a cascade of consequential events it ended up breaking the scenario. It also completely changed the tone of the game from Mad Max to Mad Magazine to the point we wound up the game and never played it again as it wasn't really doing it for us (we already had Paranoia for that sort of silliness). That was not a fault of the rules it was applying the rules without considering the consequences.

 

[Sigtrygg]

Then the issue is the referee allowing the characters to acquire the +6 bonus in the first place - just like it is in any other RPG.

It's not the referee allowing it, it is the game rules allowing in. If the referee bans it under rule zero you are not playing bu the rules as written.

 

[Sigtrygg]

The Effect rule is what saves things for me. There's your distinction between +6 with a low roll and +6 with a high roll.

This is similar to how MT handled rolling higher than the target number, but in MT it was for every multiple of 2 above the target number. Playing this for a bit showed this to be too low a threshold, again due to the effect of having bonus DMs (although not as high as MgT bonus DM stacking)

One of the reasons for the task system in DGP/MT was to remove the need to meticulously sum all possible DMs, the refere was just meant to raise the difficulty target number. MgT is in this curious state where it has three mechanism - stack DMs, assign difficulty levels, then boon/bane on top of that.

I would remove DM stacking and make a lot more of the boon/bane mechanic (and have done so successfully now for nearly a decade since 2016 edition). We also playtested a variant where you can have more than one boon or bane dice, but that skews results way more than we were happy with.

Removing effect as a point by point and replacing it with a rule of 4 is something else that has been tested, For every 4 over the target number you gain a benefit - extra damage die, halved armour, boon on next roll that sort of thing.

 

[Sigtrygg]

I feel like people in this discussion are missing the forest for the trees.

Augments, mods, high TL equipment and all the rest don’t exist to create auto-successes. They exist so the PCs can do the occasional Impossible Thing.

They also more than trivialise standard and difficult tasks)

That’s why there’s a scale of target numbers up to Impossible. If you let your PCs use that stuff when they don’t really need it, well yeah, they will always succeed. At that point you shouldn’t even ask for a roll. And that’s probably a boring story.

This leads to difficulty creep where suddenly the routine task of skimming from a gas giant requiring no dice roll at all becomes a standard task due to wanting to make things fun... players are only having fun if they are rolling dice.

2D6 isn’t really broken but it is limited.

Once you have gone beyond the limits it can be considered broken.

GURPS did 3d6 but 4D6 with Average TN=16 is interesting… you’re really wanting that +1 or +2 DM when you’re lacking decent gear but then Formidable 20+ and Impossible 22+ are exactly what they say without the proper gear for the job. And that’s probably a very exciting story.

The more d6s you are rolling then the more the standard deviation and the bell curve matters. How often do you see players roll a natural 12? How often do you see them roll a natural 12 followed immediately by another natural 12.

Refs - use the Difficulty scale! Not everything is nor should be 8+ !!!

Which means you no longer need the lower target numbers, so for the street level superhero game your standard target number becomes 10 or 12, and you are no longer playing the game with a standard target number of 8+

 

[Endie]

Every item in the game is optional. If you apply all optional positives and ignore all optional negatives you will get a different game. Not a better game or a worse game just a different one. No one is judging you for wanting to do that. If you do that and it breaks the game it is not however an indication the game is broken. Rule zero applies.


If you and your players want to have everyone with +6 to skills that is fine. It is however a little implausible to have characters who can do a routine thing today unable to do it tomorrow under the exact same circumstances because of "reasons".

I am saying if you want your players to be Dan Dare pilot of the future who can hot wire a Venusian Cruiser with a paperclip, it seems a little implausible (and pointless) that Dan might crash land his own shuttle unless the local circumstances are unfavourable enough to disadvantage him (he is rushed, there is bad weather, Jimbo installed the avionics incorrectly etc.).

We make rolls when the results are interesting. Forcing players to make rolls for routine thing that might randomly derail the story is not interesting to every table. Some tables like chaos and they may wish to check even routine activity just in case it throws up a hilarious incident where "Kryten slips on a bar of soap". Others will want to focus on their characters impacting the universe but may wish to throw in the odd quirky session "That day Dave went to the fish market" to keep things fresh.


Currently all this is already possible in the RAW. Your argument seems to be that we should change the rules because they break your game.

I had a character that died in the first 10 minutes of our first session of Aftermath when getting out of a bunk bed. The referee wanting to show case the skill system and made me roll to see if I fell (falling while climbing was in the rules). I slipped, fell 6 feet, rolled on a table and critically failed braking my neck. It was memorable (and hilarious at the time) and we talked about it for weeks afterwards, but replacing my character proved to be an issue within the structure of that nights session (it took about an hour to roll up a character and I would be out of the game all that time). Due to a cascade of consequential events it ended up breaking the scenario. It also completely changed the tone of the game from Mad Max to Mad Magazine to the point we wound it up the game and never played it again as it wasn't really doing it for us (we already had Paranoia for that sort of silliness). That was not a fault of the rules it was applying the rules without considering the consequences.

Well, your anecdotes and bad faith ad hominems are fine and all but you literally describe why you believe the game doesn’t work as written in your own attempt to claim otherwise!

 

[Sigtrygg]

No-one was arguing the probabilities. Anyone who has ever played an RPG game using dice probably knows this (and likely spent hours in their room generating probability analyses to min-max their character or WH40K squad). I was taught it in secondary school. However as you kept insisting there were 12 possible "outcomes" whereas even by your own argument there are only 11 led me to assume we were misunderstanding each other.

Mea culpa, yes 11. I should learn to count the results in the tables I produced.

There seems to be confusion between outcomes and events. Outcomes in probability are all the possible combinations of dice results. There are 36 outcomes just as there are for a d66 each with exactly the same probability (1/36). What you decide to do with the numbers in your experiment does not affect the probabilities of those numbers occurring. Events are the number of results of interest from your outcomes. There are 11 events if you add the dice together and group them by score each event may have a different probability (as is the case with 2d6).

What you are describing are not pobabilites.

I was hoping you would be able to see that I was not actually changing the game mechanic, simply restructuring how it was set out to simplify the argument. There is no actual difference in probabilities of any specific result between throwing three dice and discarding one of them vs rolling two dice and re-rolling one of the two and choosing the result from that dice you prefer. Both use 3 dice rolls, both end up choosing the best (or worse) two of the three. If I had said you had to discard one dice before re-rolling it and take the new result then the mechanics would not be equivalent.

There is a completely different probabilty outcome between rolling 3 dice and choosing two and rolling 2 dice and re-rolling one of them. They are completely different game systems.
Go and study it and you will find the difference, or if you want I can post at length on it.

If you cannot see that then the conversation is indeed pointless.

I can see that you do not understand that your system and the mechanic in the book have fundamentally different probability outcomes. So you are correct, you are discussing a house rule that does not agree with the rules as written.

Perhaps you will be able to accept that a 2d6 roll in traveller is sometimes practically a 3d6 roll and even by your argument has more that 11 outcomes even though only 11 different numbers can be generated.

11 results, it was very late last night, I will agree with but the rest of the statement makes no sense, a 2d roll can not be a 3d roll.

Further due to other mechanics in the game (like Deadly) not all D6 rolls are equal. In some cases a roll of snake eyes is always a fail regardless of what number you finally end up with. A flat statistical model of the sum of 2 dice shorn of context does not describe that complexity.

This is correct, they are different mechanics to the boon/bane mechanic. To determine the probability of treating a 1 or 2 as a 3 etc is a complicating factor, but is still different to rolling 3 dice, you do not re-roll the dice that result ion 1 or 2, you just read them as 3, thus if you are rolling 3d for damage say you still have 216 possibilities, but the part of the bell curve you read is shifted due to 1s and 2s now being treated as 3s.

Remember this discussion had largely devolved into "2d6 Bad d20 Good". I was merely trying to point out that the 2d6 system is more than a badly structured percentile system and is capable of nuance despite the supposed limitation of only being able to generate 11 numbers on 2d6.

Oh I agree. Using two differnt coloured ddice 0not necessary but makes things easier) you can extract much more potential from the roll

you can read them as a d66, a 2d6, two independent d6, one or both could be read as a d2 or d3. Lots of potential. The only issue I have with 2d6 and the MgT task system stacking DMs making the standard 8+ irrelevant.

A d20 is not much differnt to rolling 2d10 as percentile, but again the 2d10 roll can be used in lots of ways the d29 can't. But that is the nature of the D&D system, different polyhedrals for different tasks.

In reality MGT2 is not a pure 2d6 system it is an d6 based system (as it doesn't use polyhedral dice), but sometimes you will be rolling 2 dice for a skill, sometimes 3 dice and sometimes the values on those dice will change for specific reasons and you will probably be adding to the rolled value. 2d6 system is a heritage term.

I agree.

For d20 a +1 DM always moves the probability of achieving a particular total by 5% regardless of that target number (notwithstanding the boundary conditions of 1 and 20).
For the 2d6 system a +1 DM can changes the probability by a different amount depending on what your target number was.

And this is where stacking DMs throws the system out of whack. A +6 bonus on a d20 roll grants a bonus of 30%, a +6 bonus on a 2d6 roll with a standard target number of 8+ grants a success chance of 100%

5th edition also has an advantage/disadvantage mechanic. It doesn't make the D20 any less swingy though so it's effects can be . It also allows adding other types of dice to the d20 for specific effects. This provides more variety but combinations of DMs can be just as impactful as adding adding +6 to a 2d6 roll. If you are a reasonable level character (equivalent to a Traveller 5-term) using a magic sword+2 with +2 or a high stat, and expertise plus a few extras for bardic inspiration and bless (equivalent to special equipment) you could easily be adding +15 to your skill roll. We don't know what 5th Edition Traveller will be doing, but if it is not using equivalent mechanics then it isn't really a 5th edition port.

I agree with this analysis.

2d6 reinforces (but does not overwhelm) the average, but it does mean that if you too generous with your positive DMs, not diligent in enforcing the conditions required for those DM's to apply or not applying appropriate negative DMs your game can "run hot".

The game tells you what is allowed. if you restrict the players in equipment or artificially asign higher difficulty numbers to every task you are no longer playing the game as written.

Players that want to "win" will focus on their characters strengths and avoid situations where they are weak. There are fewer worthwhile stories there as there is less challenge and it is for the referee to put them outside their comfort zone (while allowing them to shine when they are the right person for the job).

I agree.

Traveller (and contemporary games like 1st edition D&D) used to be about dealing with the challenge presented. This started at character generation as you got very little choice and even those choices were subject to random fluctuation. Referees were expected to be impartial but fair. Survival was a challenge and when you succeeded in a mission you felt good. When you failed, you sucked it up and endured the consequences ("Dingo is too wounded to continue, we'll have to leave him here and go on alone").

This is pretty much how I still run games and my players play them.

Modern sensibilities seem to be focussed on what challenges the players would like to face, the referee is expected to support their choices. Success is expected and failure is shorn of real consequence ("We'll have long rest and get all our HP and abilities back"). That changes the nature of the game far more than what dice mechanic you happen to use.

I don't run games like that, and players with that mindset do not stay long at the table.

 

[swordtart]

What you are describing are not pobabilites.

No they are the fundamental terms when discussing probabilities. The actual numbers probabilities you get are not in dispute and you have already posted them so I see no need to discuss them.

There is a completely different probabilty outcome between rolling 3 dice and choosing two and rolling 2 dice and re-rolling one of them. They are completely different game systems.
Go and study it and you will find the difference, or if you want I can post at length on it.
I can see that you do not understand that your system and the mechanic in the book have fundamentally different probability outcomes. So you are correct, you are discussing a house rule that does not agree with the rules as written.

This is where I disagree and as this constitutes a matter of fact rather than opinion I will take you up on your offer of an explanation if I have misunderstood how the probability pans out I will gladly take the lesson. I hope you will do likewise.

I suspect you will realise that the difference commonly cited with 3d6 drop lowest vs 2d6 and re-roll is because many games force you to take the new result. As stated in my method you can retain either result for the re-rolled die, either the new result or the original roll.

In both methods you are rolling 3 dice. In both methods you are choosing highest two. In my method you are simply delaying the roll of the 3rd die to avoid nugatory rolls. For example if the original roll is double 6 then a re-roll is unnecessary as you cannot get better. If you only need 6+ and effect is irrelevant for that particular check then if you got 6 and 1 you still wouldn't bother re-rolling. If you need 6+ and effect is important and you roll 3 and 3 you would re-roll. If the re-roll gave you 4 you would keep it, but if it rolled less than 3 you would keep the original 3.

11 results, it was very late last night, I will agree with but the rest of the statement makes no sense, a 2d roll can not be a 3d roll.

I meant a skill check is a 2d6 check except when it isn't (but we still refer to it ass a 2d6 skill system). If it is no longer a 2d6 check then the probabilities of 2d6 no longer apply.

This is correct, they are different mechanics to the boon/bane mechanic. To determine the probability of treating a 1 or 2 as a 3 etc is a complicating factor, but is still different to rolling 3 dice, you do not re-roll the dice that result ion 1 or 2, you just read them as 3, thus if you are rolling 3d for damage say you still have 216 possibilities, but the part of the bell curve you read is shifted due to 1s and 2s now being treated as 3s.

Now you are saying 3d6 gives 216 possibilities. There are possible 216 outcomes. Under the argument you put forward for the 2d6 roll you would say there are only 16 outcomes (total value = 3 to 18). I would call these events. However since the triggering condition for the boost up to 3 is not simple (as for example as adding 2 to whatever you roll) it does not shift you up the probability curve, it changes the shape of the curve. 18 is still the maximum, but now 9 rather than 3 is the minimum. There are no longer 16 events possible, there are only 12. The chance of getting 9 is now vastly greater than it was previously but the chance of getting 18 is the same and the chance of getting 3 is nil.

The game tells you what is allowed. if you restrict the players in equipment or artificially asign higher difficulty numbers to every task you are no longer playing the game as written.

The game tells you to tune the game to your tables play style. The skills give suggestions for difficulties for particular skill uses, and the rules give generic guidance but it does not give a comprehensive list of difficulties for every task in every situation. It also gives inconsistent guidance e.g. on CRC p60 plotting jump calculations is listed as an exemplar of an average task. Under the actual Astrogation skills CRC p65 it tells us it is an Easy task.

I am also not assigning higher difficulty levels to every task. I am quite happy that a routine task is routine and needs 6+. If a player has +6 due to their skill, characteristic modifier and specialist equipment then I don't see any reason to ask for a check. That character is clearly only going to fail in exceptional circumstances. If there are no such circumstances then you are cheating the player by artificially introducing a risk of failure. If the circumstances exist then it might apply a negative modifier, shift the difficulty or create a bane. Which way it goes is and always has beena referee (or scenario designers) choice.

 

[Sigtrygg]

This is where I disagree and as this constitutes a matter of fact rather than opinion I will take you up on your offer of an explanation if I have misunderstood how the probability pans out I will gladly take the lesson. I hope you will do likewise.

Ok but it will be long and read like a maths lesson
No sleight is intended but you don't understand probability if you thing roll 3d drop a die and roll 2d then re roll a die are the same.

I suspect you will realise that the difference commonly cited with 3d6 drop lowest vs 2d6 and re-roll is because many games force you to take the new result. As stated in my method you can retain either result for the re-rolled die, either the new result or the original roll.

They are different mechanics with different probabilities.

If effect you are simply delaying the roll of the 3rd D6 so to avoid nugatory rolls. For example if the original roll is double 6 then a re-roll is unnecessary as you cannot get better. If you only need 6+ and effect is irrelevant for that particular check then if you got 6 and 1 you still wouldn't bother re-rolling.

No, when you roll 3d you have 216 "outcomes" with 16 possible numerical results.

When you roll 2d you have 36 possible "outcomes" with 11 possible numerical results. if you then reroll a dice you have a 1 in 6 chance of getting any number.

I meant a skill check is a 2d6 check except when it isn't (but we still refer to it ass a 2d6 skill system). If it is no longer a 2d6 check then the probabilities of 2d6 no longer apply.

I am being a bit thick, I do not follow what you mean here.

Now you are saying 3d6 gives 216 possibilities. There are possible 216 outcomes. Under the argument you put forward for the 2d6 roll you would say there are only 16 outcomes (total value = 3 to 18). I would call these events.

I changed my terms to match yours, I will refrain from doing so.

However since the triggering condition for the boost up to 3 is not simple (as for example as adding 2 to whatever you roll) it does not shift you up the probability curve, it changes the shape of the curve.

This is wrong. The shape of the curve does not change, you just move it along the x axis. A 2 becomes a 4 all the way up to a 12 becoming a 14, on the x axis, but the probability curve remains the same, the percentage chance of getting any particular die result remains the same.

The reroll by your own admission changes the shape of the probability graph and as a result the two mechanics are very differnt, you prove my argument for me. Do you still need the explanation?

18 is still the maximum, but now 9 rather than 3 is the minimum. There are no longer 16 events possible, there are only 12. The chance of getting 9 is now vastly greater than it was previously but the chance of getting 18 is the same and the chance of getting 3 is nil.

You have changed the probability distribution so it is no longer the same mechanic.

The game tells you to tune the game to your tables play style. The skills give suggestions for difficulties for particular skill uses, and the rules give generic guidance but it does not give a comprehensive list of difficulties for every task in every situation. It also gives inconsistent guidance e.g. on CRC p60 plotting jump calculations is listed as an exemplar of an average task. Under the actual Astrogation skills CRC p65 it tells us it is an Easy task.

I am also not assigning higher difficulty levels to every task. I am quite happy that a routine task is routine and needs 6+. If a player has +6 due to their skill, characteristic modifier and specialist equipment then I don't see any reason to ask for a check. That character is clearly only going to fail in exceptional circumstances. If there are no such circumstances then you are cheating the player by artificially introducing a risk of failure. If the circumstances exist then it might apply a negative modifier, shift the difficulty or create a bane. Which way it goes is and always has beena referee (or scenario designers) choice.

The "routine" name for this tyask target number is another of my bugbears, Something that is routine is trivial, average, very likely to succeed.

I prefer the term standard for the 8+ roll for this reason and work from there.

 

[swordtart]

Ok but it will be long and read like a maths lesson
No sleight is intended but you don't understand probability if you thing roll 3d drop a die and roll 2d then re roll a die are the same.

Go ahead. I'll post my own analysis and we can compare notes.

They are different mechanics with different probabilities.

No, when you roll 3d you have 216 "outcomes" with 16 possible numerical results.

Only if you are adding them together. You never add the 3 dice together. If you roll 3 and drop 1 the dropped one is not counted. You still have a range of 2-12 = 11 numerical results or events for the two you counted.

When you roll 2d you have 36 possible "outcomes" with 11 possible numerical results. if you then reroll a dice you have a 1 in 6 chance of getting any number.

The same number of numerical results (events) and 36 outcomes for the 2 dice times the 6 possible outcomes of the other die = 216 overall outcomes. Identical.

I am being a bit thick, I do not follow what you mean here.

I don't know what your issue with it is. Let's just forget it as it was just restating something we both agreed on in that "MGT2 is not a pure 2d6 system". We clearly have a more fundamental disconnect here.

I changed my terms to match yours, I will refrain from doing so.

Not my terms, those of probability theory (or at least they are in the classes I took). Check the internet and see if any reputable maths site (not reddit) says different.

This is wrong. The shape of the curve does not change, you just move it along the x axis. A 2 becomes a 4 all the way up to a 12 becoming a 14, on the x axis, but the probability curve remains the same, the percentage chance of getting any particular die result remains the same.

Ok this is way off and you are confusing yourself. What you are describing is adding 2 to the result of a 2d6 roll. That is a straight translation of +2. We were talking about a "Deadly" weapon that rolls 3 dice damage. Rolls of 1 and 2 are treated as 3. A 6 is still a 6. So the results can be 9 to 18 only. There are only 12 possible results whereas if the weapon were not deadly there would be 16 possible results 3-18.

If you now have 12 possible results but you previously had 16 your points are now in closer together and you have compressed the curve not shifted it. Previously you had 1 in 216 chance of rolling 3 which was your lowest value on the y-axis. Your next point (a total of 4) had a 3 in 216 chance and was you next lowest value. For the shape to be preserved in counting all 1-3 as 3 your chance of getting 9 would have to be 1 in 216. It isn't. As your chance of rolling 18 remains at 1 in 216 under both methods the shape has changed as there is no common transation between the lowest value and the highest value in the range.

The reroll by your own admission changes the shape of the probability graph and as a result the two mechanics are very differnt, you prove my argument for me.

No! that is misrepresenting my point. I said changing rolled 1-3 to 3 in the case of "deadly" changed the shape of the probability curve. I did not mention a change in the curve by rolling 3d6 drop one vs roll 2d6 and re-roll retaining the best result.

The "routine" name for this tyask target number is another of my bugbears, Something that is routine is trivial, average, very likely to succeed.

I prefer the term standard for the 8+ roll for this reason and work from there.

But that is a change from MGT2 RAW. Routine is routine for someone with some competence. Clearly if you don't have the competence you will be at -3 and your chance of success reduces significantly.