Gaim in P&P

[silashand]

What exactly is the problem here?

The problem is no one has proven that the Skrunnka was broken because it had hull 6 before and no one has proven that it is broken at hull 6 under the new Gaim fleet rules. I'm stating my experience is that it isn't reliably survivable enough to do what it's charged with doing as it stands now based on my experience playing with it. The arguments against returning it to hull 6 to make it slightly more survivable have so far been proven inaccurate (better weaponry than other ships, pods an auto-win, etc.). IMO this suggests that too many changes were made to the ship in the original revised list (is "original revised" an oxymoron? ).

The evidence I have asked for is playtest results proving that returning it to hull 6 is broken. So far we've only seen theoryhammer opinions that it will be. In the one example provided so far, the Urik'Hal vs the Skrunnka, the scenario was weighted heavily in favor of the Skrunnka (10" apart to start? Try that with a White Star -the same PL level, even the revised version- as an opponent to the Skrunnka and see how well it works out). Which BTW illustrates quite well the falacy that ships in this system can ever be balanced on a ship-by-ship basis. Some attributes are simply better all around than others, no matter the combinations. Not that they cannot be toned down to reasonable levels, but some ships in a given PL will always be better than others. My opinion is that the Skrunnka is not as good as it should be given its role and other attributes.

The problem to me appears to be that people recall how broken the fleet was before and are so paranoid about it being broken again that they are unwilling to consider *any* positive change to the fleet even if it may be needed. I think it *is* needed in this case. I am willing to be convinced otherwise, but I won't be by arguments that cannot stand up to scrutiny which so far they have not. If someone plays with the Skrunnka as part of an entire fleet over several games and shows me that my experience is an anomaly, then fine. I can abide by that. However, I haven't seen that to date. Given the apparent lack of Gaim players nowadays I'm not sure it will actually appear (they all were beaten back to their homeworld apparently ;-)).

Cheers, Gary

 

[silashand]

So timing has a lot to do with it. Get your good beam rolls early in the game, they are horribly, horribly effective.

I don't think anyone is arguing that (at least I'm not). The point is that luck in and of itself cannot be used as a basis for statistical analysis. The other factors involved, i.e. lucky/unlucky, timing, your opponent's luck/unluck with dice such as CBD, etc., none can be reliably quantified. What can be evaluated are the odds something out of the ordinary will occur and the actual capabilities of other, non-luck related factors such as speed and range. In the case of the Skrunnka, 12" range is short by most comparisons and SPD 8 is average. Hull points on the other hand are slightly higher than average which is one measure of survivability, but are mitigated by the prevalence and importance of critical hits in this game. Those are the factors that can be evaluated. Luck cannot. As much as we love it when it goes our way and hate it when it doesn't, there is no actual way to quantify the specific effect it will have in any given game or at any given moment. That's why I don't particularly like the existing beam mechanism (or crits for that matter) as they add an element of random chance that can be so outside the bounds of reasonable gameplay that it can potentially destroy a game or save it in almost one roll of the die. I don't think any game should be that susceptible to extreme random chance because it takes the fun out of actually playing sometimes, which if you think about it is why we all play the game. JMO though.

Cheers, Gary

 

[Target]

I still don't like the gaim but that is due to e-mines. Im a e-mine hater.
If i had my way the gaim would be a lot different.
No e-mines, for i'd give them photon lances, long lances (probably AP, Precise 20" & 30" ) & pulse cannons. Maybe have different type of weapon entirely Photon splash cannon, Short ranged, Weak but DD or TD. Hard to hit with but can do a lot of damage if it gets through. Kind sick of evryweapon with SAP gets precise or DD. Want a lot more varations in weapons over all fleets.
Lot's of fighters/drones which can still suicide, a hybrid heavy fighter/breaching pod.
No interceptors, the fighters can do that, cut the AF down a bit on non queens. Still lots of troops. Maybe give them a mine layin corvette, variant of the stak with abbai SA that they are lookin at. They did have mines not e-mine lauchers.
My theory is the queen ships bombard the enemy fleet while the fighters either protect the queens or strip the interceptors off the enemy for the lances. The lances cripple the opposing ships making it easier to board. The non queens zoom in & generally harass with pulse cannons.
The Hybrid breaching pod thingy has crap dogfight but AF1 & maybe a small missle. The mine layer corevette thing might lay down a screen of mines for enemy ships coming in on the queens.
This should be more fun hopefully but this is from a e-mine hater. I really want to see less of them not more. Don't mind people being able to lay mines. Mine layers & sweepers would be a cool addition.

 

[Taran]

(is "original revised" an oxymoron? ).

Cheers, Gary

I gotta say "No" but then I've played an awful lot of Warhammer Fantasy for a long time, so I'm very familiar with the term "original revised" and all variations thereof.

 

[Ripple]

Silas, you haven't debunked anything...

You say the average of a 2 AD beam is 1 to 2 hits... which I agreed with above. One hit plus a fraction we can argue over. But since we don't have fractional hits... we have to round to a whole hit. The fraction is higher than .4999 in my opinion so I round up. Same for the 4 AD arc. So the Nova comparison is dead on... and the Gaim ship is better.

You continue to have no comments on the bypassing hull and interceptors as a huge boon against high end ships.

You continue to treat beams special trait as a wash with vanilla dice. it's not. At worst you can do is roll zero hits... all systems can do this. But beams CAN roll up... this is a boon... a boost... an enhancement. You can't act like this is nothing. If you are balanced you have to pay for opportunity to exceed what is fixed for others.

Your analysis doesn't stand up... give me the Nova vs the Gaim against common hulls... 5/int and 6/int, 5/dodge... show me what your math says. Once we know if the ships balance, we can look at fleet considerations and if additional changes are needed. How about the Bimith?

On balance, you bring up a system that requires additional ships... this is true, but if you respond to the whole post... including the last line, the ship has to be balanced on a number of levels. One is ship vs ship of a similar pl.

When you get into special rules you have to go to additional levels... I can't balance stealth ship vs ship either, because it's specifically vulnerable to scouts and additional hulls interacting. But that does not mean I can give a stealth ship a huge boost because multiple ships can hurt it. What happens when he takes out the scout... all of a sudden he's just plain better than the scoutless fleet...

ie the tourney fleet all of a sudden must have a scout because one fleet had all its ships balanced assuming you'd bring one... and that it would survive some significant length of time. Right back to the 'unfun' rule because balancing fleets assumes you have counters to his defenses, when depending on the fleet you may not.

Balancing on a fleet basis only works if you know who is fighting who and can build in the assumptions that were made about the opposing fleets, and the balances are close enough that one missing piece from the opposition doesn't create the r-p-s scenario.

Your claiming that the ship needs hull 6... and you don't want to compare with other ships... what else are we to use as a basis?

Ripple

 

[skavendan]

What exactly is the problem here?

The problem is no one has proven that the Skrunnka was broken because it had hull 6 before and no one has proven that it is broken at hull 6 under the new Gaim fleet rules. I'm stating my experience is that it isn't reliably survivable enough to do what it's charged with doing as it stands now based on my experience playing with it. The arguments against returning it to hull 6 to make it slightly more survivable have so far been proven inaccurate (better weaponry than other ships, pods an auto-win, etc.). IMO this suggests that too many changes were made to the ship in the original revised list (is "original revised" an oxymoron? ).

The evidence I have asked for is playtest results proving that returning it to hull 6 is broken. So far we've only seen theoryhammer opinions that it will be. In the one example provided so far, the Urik'Hal vs the Skrunnka, the scenario was weighted heavily in favor of the Skrunnka (10" apart to start? Try that with a White Star -the same PL level, even the revised version- as an opponent to the Skrunnka and see how well it works out). Which BTW illustrates quite well the falacy that ships in this system can ever be balanced on a ship-by-ship basis. Some attributes are simply better all around than others, no matter the combinations. Not that they cannot be toned down to reasonable levels, but some ships in a given PL will always be better than others. My opinion is that the Skrunnka is not as good as it should be given its role and other attributes.

The problem to me appears to be that people recall how broken the fleet was before and are so paranoid about it being broken again that they are unwilling to consider *any* positive change to the fleet even if it may be needed. I think it *is* needed in this case. I am willing to be convinced otherwise, but I won't be by arguments that cannot stand up to scrutiny which so far they have not. If someone plays with the Skrunnka as part of an entire fleet over several games and shows me that my experience is an anomaly, then fine. I can abide by that. However, I haven't seen that to date. Given the apparent lack of Gaim players nowadays I'm not sure it will actually appear (they all were beaten back to their homeworld apparently ;-)).

Cheers, Gary

I would want to see battle reports in detail that support you saying it needs better. As carriers go it is damn good in my opinion. Theory talking doesn't work in games averages doesn't always pan out. I mean I once played the Minbari and failed about 20 stealth rolls in a row... I'd scouted the buggers too.

 

[stepan.razin]

I play an all-beam fleet and yes, they *can* be as good as you say. However, they can also be horribly ineffective. I've had just as many games where I rolled 20+ dice and got 2 hits as I have where I've rolled 2 dice and gotten 20 hits. However, they average out over a game usually. One volley they will do nothing, one they will do everything. Given timing and everything else, i.e. when you need them to hit they may not, I have to stand by my claim that they are not nearly as good as you claim them to be.

OK; lets assume that during the course of a game, you will get one 20+ beam hits roll, and a whole bunch of 0 or 1 hits, from your 6AD beam. I think we can all agree, that is how most games go.

If you get the 20+ hits at the start of the game, the damage you do will probably kill a high-PL ship, taking it out of the game, no return fire form it. So you are more likely to win.
If you get your lucky hit it at the end of the game, you are down to only 2 ships left yourself while your enemy hasn't lost anything, even if you kill his flagship he will probably finish you off quite easily.

So timing has a lot to do with it. Get your good beam rolls early in the game, they are horribly, horribly effective.

Well said, I would also like to add that there is a huge phycological factor to beams... Depending on the giving or a receiving end of a beam roll, you will always remember a runaway beam roll or a complete wiff at a critical moment, while your opponent we'll remember the opposite

 

[stepan.razin]

You continue to treat beams special trait as a wash with vanilla dice. it's not. At worst you can do is roll zero hits... all systems can do this. But beams CAN roll up... this is a boon... a boost... an enhancement. You can't act like this is nothing. If you are balanced you have to pay for opportunity to exceed what is fixed for others.

Beams can roll up, while a SAP weapons can be made twin-linked by a scout (cant ignore that anymore then you can the roll up effect)

Beams probably work better against a high armour, high interceptor fleets, while SAP is better against others. (and there are a lot more other fleets out there).

 

[silashand]

EDIT: deleted. Realized I've spent far too much time posting in this thread and that I really don't care all that much about explaining the tenets of good systems design. There are plenty of books out there on the subject if folks are really interested.

Cheers, Gary

 

[silashand]

Well said, I would also like to add that there is a huge phycological factor to beams... Depending on the giving or a receiving end of a beam roll, you will always remember a runaway beam roll or a complete wiff at a critical moment, while your opponent we'll remember the opposite

Agreed. The beam psychological factor is *huge* even if it doesn't happen because when they're hot you sit there and say to yourself with each roll, "you can stop now!" On the other hand, when all the initial dice come up misses you involuntarily give a sigh of relief. And yes, we all remember only the really good and really bad rolls. The only reason we recall the extremes in ACtA is that there is always that one possibility in 10,000 that you will get 20+ hits with a triple damage beam rolling only one die. We've all seen it happen. The problem isn't that it happens so often, the issue here is that we claim Beams are so good based on those one-off results when on average you do as I posted before, a relative number of hits equal to the starting AD of the weapon. All things considered that's not great, nor is it bad. However, saying that there is the possibility of the "rollup" does not make them better, it only makes them *seem* better because it's what we remember when it occurs. Personally, if you leave the hit on 4+, but cap them at 3 rolls max you solve the problem while retaining the ability to do decent damage which is what they should do. JMO though.

Cheers, Gary

 

[silashand]

I would want to see battle reports in detail that support you saying it needs better.

As I noted in my previous response, I will try to arrange a couple games using it and post the results. Like most I don't record my games unless I have a reason to do so .

Theory talking doesn't work in games averages doesn't always pan out. I mean I once played the Minbari and failed about 20 stealth rolls in a row... I'd scouted the buggers too.

That's happened to all of us, but averages often play out over multiple games, not with one or two. They also play out in the different rolls required, i.e. you may roll heck of a lot of dodge dice on a 5+, but none when trying to hit hull 5. That doesn't make hull 5 bad, nor does it make dodge 5 good. It means the dice averaged between all rolls which is why you need to play multiple games to see what is truly broken and what is not. What I recall from my most recent Gaim results (admittedly a couple months old now as I have my new Centauri fleet ), is that the assault ship is not as good as people here claim and I have seen no other Gaim players pipe up to say what their experience is. It is entirely possible I could be wrong, but at least I have played them using the current rules (at hull 5). I'm basing my arguments on my own experience, but I may be an anomaly. I'm willing to concede that if others can also provide their own results, but so far all I have to go on is what I know personally.

Cheers, Gary

PS. For the record I still think the gunship sucks even with the minor enhancements it recieved. It just doesn't seem to be able to fit any particular role in the fleet very well and I've tried to use them in every battle as well. Maybe I'm using them incorrectly, but even as a Skirmish choice they seem pretty poor all around and with the new FAP, they will be even less useful I think. JMO though and again, I'd like to hear others' experiences with them.

 

[stepan.razin]

In the immortal words of Homer Simpson: "In this house we obey the laws of thermodynamics"... well in our case its the laws of mathematics.

 

[Greg Smith]

Doesn't work that way. A 50% chance of something happening may round up in normal math, but in probability it does not. You have an equal chance of getting 1 hit as you do 2, therefore the spread is 1-2, *not* 2.

But the spread is not 1-2. You are ignoring the roll up. The spread is 0-infinity. The average however is different and is 0.99999 recurring, per AD.

 

[Burger]

Doesn't work that way. A 50% chance of something happening may round up in normal math, but in probability it does not. You have an equal chance of getting 1 hit as you do 2, therefore the spread is 1-2, *not* 2.

But the spread is not 1-2. You are ignoring the roll up. The spread is 0-infinity. The average however is different and is 0.99999 recurring, per AD.

You're right Greg, and 0.999999 recurring = 1, I can prove that mathematically in about 15 different ways if you like

Simplest: 1 - 0.99999 recurring = 0.00000 recurring....

 

[silashand]

Sorry, you are correct the spread is 0-infinity. My misspeak. However, I'm not ignoring the roll up. I'm calculating probabilities, in which case the odds that you will get 1 hit is the same as getting 2. This is not the same as the probability of getting 0 or 1, nor is it the same as getting 2 or 3 and so forth. Probability says the odds of you getting 1 hit on 2 dice are 75%, i.e. more than half. Since the value leans toward one side it is fair to assume you will get 1, thus you can round up as your mathematical proof shows. However, in the next step the odds are exactly 50% that you will get another hit. The odds of the number of hits staying at 1 or progressing to 2 are equal. That absolutely does not round up to 2 as you are proposing. Probability does not round at values of .5. Math uses rounding to calculate averages, but it is the outcome, i.e. whether the die comes up a 4+ or not, that is useful in calculating whether a beam weapon will generate more than a certain number of hits. If you cannot with any certainty show that you will get 1 more often than 2 or 2 more often than 1, then rounding only introduces bias (i.e. inaccuracy) into the simulation. This is easily proved by rolling only a single die. Short of weighting the die itself, there is no mathematical proof that it will ever result in 1-3 more often than it comes up 4-6. It just cannot happen.

You're right Greg, and 0.999999 recurring = 1, I can prove that mathematically in about 15 different ways if you like

.5 (50%) is not recurring and does not equal 1, nor does it equal 0. That is the probability value of going from 1 to 2 hits on a 4+. Since there is an equal chance of reaching each conculsion, rounding up is inappropriate since it cannot be proven true a majority of the time, nor can rounding down.

Cheers, Gary

 

[skavendan]

I seem to document everything these days feel like some kind of lawyer wait are lawyers evil enough to be in the Psi Corp mmmm

 

[Greg Smith]

But the possibility of 3 or more hits is not mathematically insignificant, nor incalculable.

You are basing you calculation on there being an equal probablilty of 1 hit and 2 hits, but ignoring the possibility of 0 hits or 3 hits or 4 hits or 57 hits.

 

[silashand]

But the possibility of 3 or more hits is not mathematically insignificant, nor incalculable.

You are basing you calculation on there being an equal probablilty of 1 hit and 2 hits, but ignoring the possibility of 0 hits or 3 hits or 4 hits or 57 hits.

The possibility of 0 hits on 2 dice is exactly the same as getting 3 hits, i.e. 25%. While further hits are a possibility, when playing the odds they are indeed insignificant, especially when you get down to rolling only 1 die which will happen 75% of the time on 2 dice. Relying on a 12.5% chance that you will get 4 hits is to me unsound. A 6.25% chance that you will get 5 even moreso. I know some people like to play the lottery, and some do actually win (unfortunately, not me :-(), the chances that it will happen are remote enough to be of little use. That is the case here. When the end bounds of a range is infinity as you have demonstrated, the value of averages goes down proportionately the more times you roll the die. I know we all like to remember that time so and so rolled a 4+ seventeen times in a row on one die, but you have to remember, the odds of that happening are 1 in 131,072, or .0007629%. You *really* want to use that in statistical analysis? Standard practice is to discard the top and bottom 10% anyway and while these instances may be fun to watch, they are hardly useful as examples or justification.

Cheers, Gary

 

[silashand]

I seem to document everything these days feel like some kind of lawyer wait are lawyers evil enough to be in the Psi Corp mmmm

I quit that. As a systems engineer I've had to document everything all the time anyway. I prefer not to have to do so in my off time as well. Thank God my current job makes other people do most of that now. All I have to do is fix their problems :lol:.

Cheers, Gary

 

[CZuschlag]

Standard practice is to discard the top and bottom 10% anyway and while these instances may be fun to watch, they are hardly useful as examples or justification.

As I QA engineer, I can't let that one go by! There's a whole major practice out there in Quality Control that, if you know how to do it, is worth a ton of money called "Six Sigma".

For those of you not versed in the incredibly obscure arcana that is statistics, sigma is that mathematical notation for the standard deviation of the mean. Six sigma means that even if your operational process is so out-of-whack that you're 6 standard deviations of bad process out there due to events --- bad parts shipment, power outages, invoice system getting hammered, whatever --- you still deliver a quality product. It is the new IT quality standard.

Another short way of saying this is it's only OK to throw out roughly 0.001% of a process.

That's a 4 orders of magnitude different from the claim above!

The claim that throwing out the top and bottom 10% is a standard process is ridiculous!

I know this is a game, and not a industrial product. But, given the above, consistency within 1 Sigma shouldn't be too much to ask for!