How Many Points Before Marvel Is Competitive?

[Elginb]

Marvel Units are hugely powerful, but because they cost so much, armies that feature one or more of them often can't fill out the rest of the army with units that make up for their weak spots. With the possible exceptions of Venom and Spider-man, I think it's hard to build a competitive Marvel-inclusive army for 500 points or less when investing so many points in just one figure.

So I'm curious what your opinions are on what point value allows for Marvel-inclusive armies to become competitive against Classic 'Scape armies-- how many points do you need to fills the gaps left by your favorite Marvel Unit? How many points do you need to get multiple attacks, glyph grabbers, meat shields?

[The Silver Surfer]

I think 600 points is just about perfect. It allows for at least 1 slightly more expensive marvel with, or you can use several cheaper ones, or whatever. And obviously, it still allows for the squads and other heroes you may want.

[Sherman Davies]

I agree, 600-700 points is about right for mixed games.

Edit: This thread should probably be moved to Marvel Discussion.

[ElvenEnvy]

Captain America can be worked into a 400point army if the map is good, and 500 point army pretty easily.
Ironman can go in a 600 point army decently.
Hulk takes about 800 points, maybe less if your opponent isn't prepaered for him.

As you can see it varies from hero to hero, but I'd agree that a game of about 650 is ideal.
If you go for multiple heroes then your looking at upwards of 700.

[dok]

With the notable exception of Thanos, who definitely becomes more valuable as points go up, I don't think it's points that hold back the use of Marvel Heroes. They simply aren't the best value for the points. If Hulk were valuable against the current metagame, that would be true at 500 points. But it's not - he gets chewed up by stingers. That doesn't suddenly stop being true if you're playing a 700 point game.

The exception is Captain America, who does see a lot of play in mixed tournaments, even 500 point events or lower. I was recently planning to use him in a 400 point event, and I don't think I would have regretted it.

Now, if you increase the points values and don't increase the hex limits, then marvel heroes (other than Captain America) gradually become more useful, yeah. But only gradually. At some point, the classic scape armies that are popular sort of break down. If you go beyond, oh, 750 or so, without boosting the hex counts, the current popular armies stop working well.

At that point, you start looking for high-value filler heroes, and Marvel figures start becoming obvious options. But I would only pick them to fill out points when limited hexes are left - not replacing key support figures that provide synergy.

Thanos is a special case, as he clearly inceases in value as the number of expected rounds increases.

[Agent Minivann]

As the previous two posts mention, Captain America is a viable choice. I might even put him ahead of Spidey and Venom. He's really dangerous vs Classicscape.

I remember Craig saying Cap and Raelin teamed was really powerful. Come to think of it, I don't think I've actually tried it.

[docweb2004]

We had discussed this Marvel thing before, and again you say that Hulk gets chewed up by Stingers.

Exactly how many Stingers does it take to defeat Hulk? 18, 24, 30?

Going by points, for Hulk's 370 points, you could have 6 sets of Stingers for 18 Stingers total.

Say, for grins, in a 700 point match, I had Hulk (370), Q9(180) and Venom(150). (This army looks pretty good by the way)

How would you combat this army?

[joe5joe7]

We had discussed this Marvel thing before, and again you say that Hulk gets chewed up by Stingers.

Exactly how many Stingers does it take to defeat Hulk? 18, 24, 30?

Going by points, for Hulk's 370 points, you could have 6 sets of Stingers for 18 Stingers total.

Say, for grins, in a 700 point match, I had Hulk (370), Q9(180) and Venom(150). (This army looks pretty good by the way)

How would you combat this army?
4th mass and Captain America, assuming unlimited start zones :p

[Elginb]

I agree, 600-700 points is about right for mixed games.

Edit: This thread should probably be moved to Marvel Discussion.

Oops-- yeah, is there a moderator who could move it? I wasn't paying attention, apparently...

Thanks

[dok]

I remember Craig saying Cap and Raelin teamed was really powerful. Come to think of it, I don't think I've actually tried it.
The 400 point army I was going to use was Cap, Raelin, Krav. It works really well.

[Agent Minivann]

I remember Craig saying Cap and Raelin teamed was really powerful. Come to think of it, I don't think I've actually tried it.
The 400 point army I was going to use was Cap, Raelin, Krav. It works really well.
That's brutal!

[dok]

We had discussed this Marvel thing before, and again you say that Hulk gets chewed up by Stingers.

Exactly how many Stingers does it take to defeat Hulk? 18, 24, 30?

Going by points, for Hulk's 370 points, you could have 6 sets of Stingers for 18 Stingers total.
It's not nearly that many.

Unfortunately, the Heroscape matchup calculator hangs up when you try to plug in more than a couple squads of commons. But I'll crank through it by hand.

Assuming the stinger player is competent enough to spread out so that Hulk can't use stomp, Hulk gets one rage smash attack per turn, and the stingers get 3 attacks of 4 if they succeed with stinger drain, and none if they fail.

According to Sisyphus's probability tables (http://www.heroscapers.com/community/showthread.php?t=16263), This means the Stingers will cause 1.566 wounds to Hulk, on average, per turn (and lose .2 stingers, on average). Hulk has 8 wounds, so this means that, on average, it takes just over 5 turns for the stingers to kill the Hulk. To be conservative, I'll call it 6 turns.

On average, the stingers will have lost 1.2 figures to their own stinger drain in that time. Hulk, for his part, starts at a 76% chance of killing a stinger, and quickly gets into the 90s as wounds pile up. Again, to be conservative, I'll say he has an 80% chance in the first turn, and 100% for each subsequent turn. So that's 5.8 stingers killed, which, when combined with the drain kills, means 7 total dead stingers.

In order to get 3 attacks on the last turn, we need to add 3 to that total. So my final answer is: 10 stingers should have better than even odds to kill the Hulk. And remember, I rounded in the Hulk's favor at every opportunity there.

10 stingers - that's 200 points worth of figures. Hulk costs 370.


Say, for grins, in a 700 point match, I had Hulk (370), Q9(180) and Venom(150). (This army looks pretty good by the way)

How would you combat this army?
Well, the nice thing about that army is that it is pretty rat-resistant. I know I sound like a broken record, but a horde of stingers would be strongly favored against that army. If you wanted to mix it up, you could use Minions of Utgar or Knights+champions. If it wasn't for Q9 I'd bring up the 4th or the 10th. 10th+Minions would work well - sic the Minions on Q9 and the 10th can easily handle the rest. Glads+Blasts is another squad army that would do the job.

[docweb2004]

We had discussed this Marvel thing before, and again you say that Hulk gets chewed up by Stingers.

Exactly how many Stingers does it take to defeat Hulk? 18, 24, 30?

Going by points, for Hulk's 370 points, you could have 6 sets of Stingers for 18 Stingers total.
It's not nearly that many.

Unfortunately, the Heroscape matchup calculator hangs up when you try to plug in more than a couple squads of commons. But I'll crank through it by hand.

Assuming the stinger player is competent enough to spread out so that Hulk can't use stomp, Hulk gets one rage smash attack per turn, and the stingers get 3 attacks of 4 if they succeed with stinger drain, and none if they fail.

According to Sisyphus's probability tables (http://www.heroscapers.com/community/showthread.php?t=16263), This means the Stingers will cause 1.566 wounds to Hulk, on average, per turn (and lose .2 stingers, on average). Hulk has 8 wounds, so this means that, on average, it takes just over 5 turns for the stingers to kill the Hulk. To be conservative, I'll call it 6 turns.

On average, the stingers will have lost 1.2 figures to their own stinger drain in that time. Hulk, for his part, starts at a 76% chance of killing a stinger, and quickly gets into the 90s as wounds pile up. Again, to be conservative, I'll say he has an 80% chance in the first turn, and 100% for each subsequent turn. So that's 5.8 stingers killed, which, when combined with the drain kills, means 7 total dead stingers.

In order to get 3 attacks on the last turn, we need to add 3 to that total. So my final answer is: 10 stingers should have better than even odds to kill the Hulk. And remember, I rounded in the Hulk's favor at every opportunity there.

10 stingers - that's 200 points worth of figures. Hulk costs 370.


Say, for grins, in a 700 point match, I had Hulk (370), Q9(180) and Venom(150). (This army looks pretty good by the way)

How would you combat this army?
Well, the nice thing about that army is that it is pretty rat-resistant. I know I sound like a broken record, but a horde of stingers would be strongly favored against that army. If you wanted to mix it up, you could use Minions of Utgar or Knights+champions. If it wasn't for Q9 I'd bring up the 4th or the 10th. 10th+Minions would work well - sic the Minions on Q9 and the 10th can easily handle the rest. Glads+Blasts is another squad army that would do the job.

Your probability tables are incorrect. I just play tested your theory twice, and Hulk beat the stingers both times. One in 8 rounds with 5 wounds, and the other in 9 rounds with 6 wounds. When the stingers start to create wounds, the stingers just get wiped out when Hulk is rolling 8-10 die. Also, keep in mind, the stingers are usually only rolling 3 die on attack, sometimes 4. Hulk rolls 6 die on defense.

Does this tie up hulk and make him weaker, absolutely. But on the other hand, Hulk would have reinforcements. What reinforcements do you have with the Stingers to combat Hulk's allies of Venom and Q9? I still say the Stinger based army with 700 points can't defeat Hulk's Army.

I'm going head to head with a friend this weekend, you tell me what army you would use with the Stingers.......will be fun...

[dok]

Assuming the stinger player is competent enough to spread out so that Hulk can't use stomp, Hulk gets one rage smash attack per turn, and the stingers get 3 attacks of 4 if they succeed with stinger drain, and none if they fail.

According to Sisyphus's probability tables (http://www.heroscapers.com/community/showthread.php?t=16263), This means the Stingers will cause 1.566 wounds to Hulk, on average, per turn (and lose .2 stingers, on average). Hulk has 8 wounds, so this means that, on average, it takes just over 5 turns for the stingers to kill the Hulk. To be conservative, I'll call it 6 turns.

On average, the stingers will have lost 1.2 figures to their own stinger drain in that time. Hulk, for his part, starts at a 76% chance of killing a stinger, and quickly gets into the 90s as wounds pile up. Again, to be conservative, I'll say he has an 80% chance in the first turn, and 100% for each subsequent turn. So that's 5.8 stingers killed, which, when combined with the drain kills, means 7 total dead stingers.

In order to get 3 attacks on the last turn, we need to add 3 to that total. So my final answer is: 10 stingers should have better than even odds to kill the Hulk. And remember, I rounded in the Hulk's favor at every opportunity there.

Your probability tables are incorrect.

The tables are not incorrect. I'm guessing that you didn't look at them, or you would have probably caught my mistake.

However, my ability to read the Incredible Hulk card is, in fact, subject to error. I thought it was 5 defense, not 6.

So, that brings the average damage per turn down from 1.566, to 1.23. That means it takes a hair over 6.5 turns to take the Hulk down... call it 7.

That means it's roughly 8 dead stingers, not 7, so you need somewhere in the neighborhood of 11 stingers, not 10, to break 50%.

I just play tested your theory twice, and Hulk beat the stingers both times.
Sample size much? Also, you're not providing much detail here: did you spread them out to avoid stomp? Did the stingers take advantage of height when possible?

One in 8 rounds with 5 wounds, and the other in 9 rounds with 6 wounds.
So... 6 wounds on the Hulk with 10 stingers? Did it occur to you to try 11 or 12? I said it was the break even, 50% point, not that stinger victory was guaranteed.

For an apples-to-apples comparison, you could try 18 stingers. What odds to you give the Hulk in that fight?

When the stingers start to create wounds, the stingers just get wiped out when Hulk is rolling 8-10 die.Yes, the Hulk will kill one stinger per turn. This is manageable. That's the point. Unlike, say, Q9 or 4th Mass, the Hulk can't kill multiple stingers in a single turn.

Also, keep in mind, the stingers are usually only rolling 3 die on attack, sometimes 4. Hulk rolls 6 die on defense.
It seems you weren't rolling for Stinger Drain. You should have been. Against heroes, it's to your advantage to roll for the drain. That increases the stingers' odds.

Try it again with 4 squads of stingers, and roll for stinger drain this time. Try it three or four times. Hulk should die more often than not.

[dok]

Now I'm really starting to wonder about the test you ran. I just wrote a quick computer simulation to test it. I get the 50% break point to be only 8 stingers - not surprising given all the ways I was rounding things in favor of the Hulk. 8 stingers, on flat ground, giving up first attack and losing initiative every round, win 51.6% of the time, with an average of just under 2 stingers surviving the conflict.

The Hulk's chance of victory drops fairly precipitously after that. By 11 stingers, he's down to under 21%. Against 18 stingers, he has less than a 2% chance of winning. On average, about 11 stingers are left after that fight.

These numbers are based on 10000 trials (battles) per number. So, my simulation ran the Hulk versus 18 stingers fight 10000 times, and the Hulk won 176 times out of 10000.

[docweb2004]

In the test battles I did, I didn't give the Stingers height advantage, and I never gave Hulk height advantage either, but I did spread them out and attempt Stinger Drain every turn; a couple times a Stinger died, and about 4-5 times it gave them 4 attack die.

As far as Hulk, I never used the Hulk Smash.

As far as realistic roles go, Hulk was only receiving a wound about every other time, and at least one Stinger was being elimated every round.

It sounds like those probability tables are backwards. It looks like a 50/50 chance with 18 Stingers versus Hulk, assuming no height advantage and not using Hulk Smash.

Now we are up to having to use 18 Stingers, that is 360 points. Who would you use to fill your other 340 points versus Hulk, Q9, and Venom?

[dok]

In the test battles I did, I didn't give the Stingers height advantage, and I never gave Hulk height advantage either, but I did spread them out and attempt Stinger Drain every turn; a couple times a Stinger died, and about 4-5 times it gave them 4 attack die.

Do'h, I forgot about the 25% chance of a 3 attack die. The pitfalls of speed coding...

REVISED RESULTS

The break-even number, where Hulk dies more often than not, is 9 stingers. Hulk died in 5043 of 10000 trials (50.4%) against 9 stingers. An average of just a hair under 2 stingers are left.

Against 10 stingers, Hulk wins 40.73% of the time.
Against 11 stingers, Hulk wins 32.25% of the time.

Against 18 stingers, Hulk wins 4.32% of the time (432 out of 10000 trials). On average, just under 10 stingers survive the fight.

Again, this is with no height advantage anywhere, and with Hulk getting first attack and always winning initiative.

As far as Hulk, I never used the Hulk Smash.
I assume you mean the stomp. The smash (i.e. boost to normal attack) is what he uses every time.

As far as realistic roles go, Hulk was only receiving a wound about every other time,
If the Hulk was really avoiding wounds that often, he was rolling exceptionally well (or the stingers were rolling exceptionally poorly).

and at least one Stinger was being elimated every round.
How would more than one Stinger get eliminated?

Now we are up to having to use 18 Stingers, that is 360 points. Who would you use to fill your other 340 points versus Hulk, Q9, and Venom?
Assuming I'm not getting any points for style, I'll just add more stingers... boring, I know. But if all I'm doing is trying to prove a point, then what do I care about style?

[The Silver Surfer]

and at least one Stinger was being elimated every round.
How would more than one Stinger get eliminated?



Note the bolded Round. Not Turn. Big difference. A round is once everyone has used all 3 order markers, a turn is a single order marker.


Oh, and by the way, I find this discussion quite entertaining, actually. I'm still waiting for the final result, though. :p

[docweb2004]

The way they lost 2 stingers in a round, was one was lost by Stinger Drain, then Hulk killed the other one on his turn. After Hulk gets 3-4 wounds, it is automatic that you lose a Stinger every round.

Ok, so you want to go with 11 sets of Stingers? for 660 points? That will take up 33 hexes. What do you want to do with your other 40 points?

By the way, I get to create and choose the map, just joking...

I will say one thing though, I can see why this would be a boring game. Just rolling dice one on one Hulk versus Stingers........I can see the viewpoint of people that don't like to include Marvel. I don't think it is because they are too powerful, just people would load up like you and it would be just a game of luck rolling dice.

[dok]

The way they lost 2 stingers in a round, was one was lost by Stinger Drain, then Hulk killed the other one on his turn. After Hulk gets 3-4 wounds, it is automatic that you lose a Stinger every round.

OK, fair enough. Based on what you've said, I'm just going to guess your two trials fall in the 16% of the time that the Hulk would win 2/2 fights against 10 stingers. 16% is unlikely, but it sounds like you were doing things right, and something that happens 16% of the time will, occasionally, happen.

Ok, so you want to go with 11 sets of Stingers? for 660 points? That will take up 33 hexes. What do you want to do with your other 40 points?
Ugh. Can I just take two more stingers, for 35 hexes of fun? Also, could you pass the cyanide? That sounds like the least fun battle ever.

I will say one thing though, I can see why this would be a boring game. Just rolling dice one on one Hulk versus Stingers........I can see the viewpoint of people that don't like to include Marvel. I don't think it is because they are too powerful, just people would load up like you and it would be just a game of luck rolling dice.
Sort of. Big heroes do tend to reduce the tactical element of the game a bit. Still, I think most people's objections to Marvel are for other reasons.

To be clear, I think you can design much more varied, interesting armies that would match up well against the three heroes you picked. I could make an interesting army with Braxas, Warriors of Ashra, Minions of Utgar, Jotun, Krug... there's lots of variety I could build on. The reason I went with Stingers was because they are one of the only units that match up reasonably well against all three of those heroes and because, again, I was just trying to prove a point.

Also, that stinger army is hardly my pick for the strongest 660 point army. It would probably get demolished by a combination of Deathreavers and Stingers, and would get beaten even worse if that army threw in some more support figures. A pile of 4th Mass would easily defeat those stingers. Knights and Champions would cut right through them. And so on.

[docweb2004]

Just one last note, just to follow your thread, if Hulk was on a team supported by 4th Mass, Stingers, Deathweavers, or Knights and Champions, it would effective off-set your advantage of having 35 Stingers!

Could have 330 points of 4th Mass or Deathweavers to support him...

lol

Hey, thanks for the exchange, and you are right, it would make one boring game, but! I am going to play it out in one battle just to see how Hulk, Q9 and Venom would do against a massive group of Stingers....

[dok]

Just one last note, just to follow your thread, if Hulk was on a team supported by 4th Mass, Stingers, Deathweavers, or Knights and Champions, it would effective off-set your advantage of having 35 Stingers!
Yes, now you are getting it! Absolutely. If you mixed a bunch of rats&stingers in your army, or knights&champions, or glads&blasts, then yes, my all-stinger army stops looking so good. And then I have to start thinking about combining different sorts of figures to make my army effective against a wider range of opponents.

Now, the real question is, is Hulk ever a really great addition to an army? My argument, in a nutshell, is that there are better ways to spend 370 points. If you want a hard-hitting hero that forces your opponent to have some antihero squad options, then Hulk does the job, but so does a much cheaper option like, say, Nilfheim.

EDIT: to be clear, I'm not trying to argue that Nilfheim is as good as Hulk overall - just that he puts the same kind of pressures on your opponent, but leaves more points for you to spend on other parts of your army.

[Simpsons Scaper]

Just one last note, just to follow your thread, if Hulk was on a team supported by 4th Mass, Stingers, Deathweavers, or Knights and Champions, it would effective off-set your advantage of having 35 Stingers!
Yes, now you are getting it! Absolutely. If you mixed a bunch of rats&stingers in your army, or knights&champions, or glads&blasts, then yes, my all-stinger army stops looking so good. And then I have to start thinking about combining different sorts of figures to make my army effective against a wider range of opponents.

Now, the real question is, is Hulk ever a really great addition to an army? My argument, in a nutshell, is that there are better ways to spend 370 points. If you want a hard-hitting hero that forces your opponent to have some antihero squad options, then Hulk does the job, but so does a much cheaper option like, say, Nilfheim.

EDIT: to be clear, I'm not trying to argue that Nilfheim is as good as Hulk overall - just that he puts the same kind of pressures on your opponent, but leaves more points for you to spend on other parts of your army.
I think if you're playing 400, then don't take Hulk. If you are playing a massive 1000 point game, then he is a good choice.

[Simpsons Scaper]

Sorry for the double post, but I did some matchup calculatoring and 2 squads of Shades that don't attack will beat Hulk about 83% of the time. Braxas beats Hulk 55% of the time if she attacks, and 55% if she doesn't.

[dok]

I think if you're playing 400, then don't take Hulk. If you are playing a massive 1000 point game, then he is a good choice.
No, he's still not great; that's my point. Not unless you're playing a really low number of start zone hexes. He's not a waste of points, don't get me wrong, but there are better uses for 370 points. Just because you have more points available, doesn't mean the Hulk somehow becomes a better value.

At 1000 points/24 hexes, there's still lots of better options for your points. 1000 points and 14 hexes? OK, yeah, then the Hulk looks good.

Sorry for the double post, but I did some matchup calculatoring and 2 squads of Shades that don't attack will beat Hulk about 83% of the time. Braxas beats Hulk 55% of the time if she attacks, and 55% if she doesn't.
I'm not sure what you mean about Braxas not attacking. She can't both use her special ability and attack in the same turn.

With the Shades, were you giving the Hulk his stomp attack, or his regular attack? The Hulk would almost surely use his stomp against Shades, which helps a great deal. I am pretty sure he's better than 17%.

EDIT: ran it in the matchup calculator, and sure enough, Hulk wins slightly over 30% of the time.

Also, when Braxas opts to attack in stead of breathe acid, her odds of winning drop to a bit over 1%. Not good.

[mathguy]

Okay, I ran the matchup calculator on my home computer without the self-imposed time limit of 2.5 minutes on the server version. Here are the results:

EDIT: Note that the following table is wrong. See two posts down for the correct table.

Hulk with Rage Smash
vs N Marro Stingers using Stinger Drain:
N Stingers win %
---------------------
6 6.2%
7 11.0%
8 17.0%
9 24.0%
10 31.6%
11 39.4%

(My computer ran out of memory at N=12 because that required inverting a 7000x7000 matrix.) But the pattern suggests that it would require about 13 Stingers to get a win% above 50%.

EDIT: Given that the above table is wrong, the following paragraphs have incorrect deductions as well (although the table below is correct).

It turns out that the Stingers do better if they do not use Stinger Drain:
Hulk with Rage Smash
vs N Marro Stingers not using Stinger Drain:
N Stingers win %
---------------------
6 15.1%
7 26.8%
8 40.4%
9 53.8%
10 66.0%


So the answer is 9 stingers, if they do not use Stinger Drain [or about 13 if they do use Stinger Drain]. Or 9 of any common (with 3 per squad) that has 3 att and 3 def.

[dok]

Okay, I ran the matchup calculator on my home computer without the self-imposed time limit of 2.5 minutes on the server version. Here are the results:

Hulk with Rage Smash
vs N Marro Stingers using Stinger Drain:
N Stingers win %
---------------------
6 6.2%
7 11.0%
8 17.0%
9 24.0%
10 31.6%
11 39.4%

(My computer ran out of memory at N=12 because that required inverting a 7000x7000 matrix.) But the pattern suggests that it would require about 13 Stingers to get a win% above 50%.
Hmm... to be frank, I don't buy this. As I said, I wrote a simulation that ran the fight through 10,000 times, using random number generators for dice rolls, and cranking through the whole fight. I got drastically different numbers than these. I could easily run 100,000 trials, as the simulation runs very fast, but the results converge close enough to the steady state after 1000 or so, so there's no point. I'm happy to share my code if you would like to see it.

As I said before:

The break-even number, where Hulk dies more often than not, is 9 stingers. Hulk died in 5043 of 10000 trials (50.4%) against 9 stingers. An average of just a hair under 2 stingers are left.

Against 10 stingers, Hulk wins 40.73% of the time.
Against 11 stingers, Hulk wins 32.25% of the time.

Against 18 stingers, Hulk wins 4.32% of the time (432 out of 10000 trials). On average, just under 10 stingers survive the fight.

It turns out that the Stingers do better if they do not use Stinger Drain:
Hulk with Rage Smash
vs N Marro Stingers not using Stinger Drain:
N Stingers win %
---------------------
6 15.1%
7 26.8%
8 40.4%
9 53.8%
10 66.0%


So the answer is 9 stingers, if they do not use Stinger Drain [or about 13 if they do use Stinger Drain]. Or 9 of any common (with 3 per squad) that has 3 att and 3 def.
This strongly suggests to me that something is wrong with the calculation. It's very hard to believe that against a high-life figure with a defense of 6, stinger drain is a bad play.

Just step back and think about it for a second. With Stinger Drain, you do an average of (.55*1.78 + .25*1.011) = 1.23 wounds per turn, against only 1.011 without. That's a difference of .22 wounds, against a .2 chance of losing a stinger. These are pretty close numbers. So intuitively, once the number of stingers passes the number of wounds Hulk has (eight), the drain should be, at the least, roughly a break-even play.

I hate to blindly hate on your code like this, but I strongly suspect something is off.

[mathguy]

Hmm... to be frank, I don't buy this. As I said, I wrote a simulation that ran the fight through 10,000 times, using random number generators for dice rolls, and cranking through the whole fight. I got drastically different numbers than these. I could easily run 100,000 trials, as the simulation runs very fast, but the results converge close enough to the steady state after 1000 or so, so there's no point. I'm happy to share my code if you would like to see it.

As I said before:

The break-even number, where Hulk dies more often than not, is 9 stingers. Hulk died in 5043 of 10000 trials (50.4%) against 9 stingers. An average of just a hair under 2 stingers are left.

Against 10 stingers, Hulk wins 40.73% of the time.
Against 11 stingers, Hulk wins 32.25% of the time.

Against 18 stingers, Hulk wins 4.32% of the time (432 out of 10000 trials). On average, just under 10 stingers survive the fight.

Yes, it turns out I had a dumb error in my code where I forgot to increase the attack dice of a successfull Stinger Drain roll. :( D'oh! So here are the correct numbers:

Hulk with Rage Smash
vs N Marro Stingers using Stinger Drain:
N Stingers win %
---------------------
6 21.97%
7 32.59%
8 43.34%
9 53.49%
10 62.66%
11 70.62%
---------------------

So we agree that 9 Stingers is the number of Stingers to go above 50% in a battle with the Hulk. :)

Our numbers are off by a couple of percentage points.
Hmmm, did you take into account of the initiative roll every 3 turns?
I can't think of anything else. I suppose it could also be cumulative round-off errors if you hard-coded probabilities into your program using only a couple of decimal places?

...
I hate to blindly hate on your code like this, but I strongly suspect something is off.
Thanks for pointing out that there must be some coding error. It was my fault for not having enough time to read through the whole thread carefully to see that I was way off from your numbers (in general, simulations give pretty close numbers to the "exact theoretical answer").

BTW I did a theoretical calculation and the average number of TURNS it takes a group of Stingers (sufficiently many to replenish those lost to a bad Stinger Drain) to kill a non-retaliating Hulk (8 Life, 6 Def) is 7.28.
Note that this is not just 8/1.23=6.50 because the 8/1.23 estimate would not take into account of "overkilling". The 7.28 number was arrived at by calculating the exact theoretical probability that it takes exactly t turns to kill a non-retaliating Hulk (the probability percentages for t=1 to 20 are: 0.1, 1.8, 5.9, 10.3, 13.2, 14., 13.2, 11.3, 9., 6.8, 4.9, 3.4, 2.3, 1.5, 0.9, 0.6, 0.4, 0.2, 0.1, 0.1).

[Jexik]

I'm glad this got cleared up! (The Stinger Drain thing). I remember running 9 Stingers against Q9 in the matchup calculator before, and getting about 50% chance without drain, and 33% with, and that always felt way off, but I forgot to report the bug. Sorry.

[mathguy]

I'm glad this got cleared up! (The Stinger Drain thing). I remember running 9 Stingers against Q9 in the matchup calculator before, and getting about 50% chance without drain, and 33% with, and that always felt way off, but I forgot to report the bug. Sorry.

Don't be shy next time telling me that something seems not right. :)

And I must admit that when I got the initial (incorrect) results. I thought that it seemed suspicious, but it was also conceivable given that there so many parameters going on, but again it was my fault for not having enough time.

I guess for Stingers, the Drain in general is beneficial against high defense opponents and not as good against really low defense.

[mathguy]

Looking at the tables of win probabilities again:
Hulk with Rage Smash
vs N Marro Stingers using Stinger Drain:
N Stingers win %
---------------------
6 21.97%
7 32.59%
8 43.34%
9 53.49%
10 62.66%
11 70.62%
---------------------
Hulk with Rage Smash
vs N Marro Stingers not using Stinger Drain:
N Stingers win %
---------------------
6 15.09%
7 26.84%
8 40.40%
9 53.84%
10 65.96%
11 75.88
---------------------
we see that once the Marro Stingers have the upper hand (at 9 Stingers with either strategy), the Stingers actually have a higher win probability if they do not use Stinger Drain. Here is a plausible explanation: Once the Stingers have the advantage, then it becomes more significant that a failed Stinger Drain is the main way the Hulk can win.

We can also try to look at it this way.
With Stinger Drain, on average it takes 7.3 turns to kill Hulk (this assumes either Hulk does not retaliate or that you have an essentially unlimited supply of Stingers).
With no Stinger Drain, on average it takes 8.5 turns to kill Hulk.
This roughly explains why about 9 Stingers are needed in either strategy: by approximating with the assumption that the Hulk kills one Stinger per turn, then you would need at least 8.5 Stingers with the no-Stinger-Drain strategy (we assume the fact that the Hulk actually kills slightly less than 1 Stinger per turn and the fact that towards the end the Stingers don't have a full squad cancel each other out for the purposes of this approximation). And with Stinger Drain, you need an extra 0.2 Stingers per turn to make up for those lost due to a bad Stinger Drain roll and so you would need at least 7.3 +0.2(7.3) = 8.76 Stingers in the use-Stinger-Drain strategy. This could explain why once we get to 9 Stingers that the win probability for the no-Stinger-Drain strategy is higher.

If your goal is to maximize the win probability, then at 9 Stingers, you should not use Stinger Drain (though only a 0.31% increase in probability). However, if your goal is to kill the Hulk as quickly as you can, then you should use Stinger Drain (kills Hulk approximately one turn quicker). Furthermore, if your goal is to have as many surviving Stingers as possible, then you should not use Stinger Drain.

This sort of shows how well-balanced the Stinger Drain ability is. Nice work, designers!

[dok]

Our numbers are off by a couple of percentage points.
Hmmm, did you take into account of the initiative roll every 3 turns?
I can't think of anything else. I suppose it could also be cumulative round-off errors if you hard-coded probabilities into your program using only a couple of decimal places?
You got it - it's the initiative rolls. I just gave Hulk initiative every turn. (It wouldn't be that hard to do initiative rolls, but I'd have to break out the attacks into subroutines then, and I'm lazy...) I'm sure that's the source of the discrepancy.

Thanks for sorting that out so quickly.

[dok]

Looking at the tables of win probabilities again:
Hulk with Rage Smash
vs N Marro Stingers using Stinger Drain:
N Stingers win %
---------------------
6 21.97%
7 32.59%
8 43.34%
9 53.49%
10 62.66%
11 70.62%
---------------------
Hulk with Rage Smash
vs N Marro Stingers not using Stinger Drain:
N Stingers win %
---------------------
6 15.09%
7 26.84%
8 40.40%
9 53.84%
10 65.96%
11 75.88
---------------------
we see that once the Marro Stingers have the upper hand (at 9 Stingers with either strategy), the Stingers actually have a higher win probability if they do not use Stinger Drain. Here is a plausible explanation: Once the Stingers have the advantage, then it becomes more significant that a failed Stinger Drain is the main way the Hulk can win.

We can also try to look at it this way.
With Stinger Drain, on average it takes 7.3 turns to kill Hulk (this assumes either Hulk does not retaliate or that you have an essentially unlimited supply of Stingers).
With no Stinger Drain, on average it takes 8.5 turns to kill Hulk.
This roughly explains why about 9 Stingers are needed in either strategy: by approximating with the assumption that the Hulk kills one Stinger per turn, then you would need at least 8.5 Stingers with the no-Stinger-Drain strategy (we assume the fact that the Hulk actually kills slightly less than 1 Stinger per turn and the fact that towards the end the Stingers don't have a full squad cancel each other out for the purposes of this approximation). And with Stinger Drain, you need an extra 0.2 Stingers per turn to make up for those lost due to a bad Stinger Drain roll and so you would need at least 7.3 +0.2(7.3) = 8.76 Stingers in the use-Stinger-Drain strategy. This could explain why once we get to 9 Stingers that the win probability for the no-Stinger-Drain strategy is higher.

If your goal is to maximize the win probability, then at 9 Stingers, you should not use Stinger Drain (though only a 0.31% increase in probability).

This is really great stuff, and it coincides well with the super-quick back of the envelope calculation I did. I concluded that using stinger drain increased average Hulk wounds/turn by .22, at a cost of the death of, on average, .2 additional stingers per turn. So, drain gives a very very slight edge on average... but since it increases variance, it's a slightly negative play once the stingers have a statistical edge.

However, if your goal is to kill the Hulk as quickly as you can, then you should use Stinger Drain (kills Hulk approximately one turn quicker). Furthermore, if your goal is to have as many surviving Stingers as possible, then you should not use Stinger Drain.
Which means that, in most real game situations, you should use stinger drain against high-defense, high-life heroes like the Hulk. Now, if it's down to 12 stingers against the Hulk at the end of the game and there's no time constraint, then you shouldn't drain. But in most real game situations, there's other figures and other things you could be doing with your order markers, and it's well worth it to drain so you can get to those things, even if it costs you, on average, a fraction of a stinger in expected value.

To look at that another way... those extra .22 wounds on the Hulk are worth over ten points per turn, while the likely drain losses cost only four points per turn. Again, this is not relevant if it's down to Hulk and stingers and there's no time constraint, but it's relevant in many other cases.

This sort of shows how well-balanced the Stinger Drain ability is. Nice work, designers!
Well put. Stinger drain is a nice option but it's not what makes the stingers good.

[dok]

I'm glad this got cleared up! (The Stinger Drain thing). I remember running 9 Stingers against Q9 in the matchup calculator before, and getting about 50% chance without drain, and 33% with, and that always felt way off, but I forgot to report the bug. Sorry.
The calculator now reports about a 5% edge with draining (gets you over 55%). Q9's 7 defense (as oppose to Hulk's 6) makes it more of a clear-cut case. Furthermore, all my above arguments about the points-per-turn benefits of drain vs. heroes apply. You can look at padlock's recent "case study (http://heroscapers.com/community/showthread.php?t=24582)" of stingers versus heroes for more thoughts along these lines.

[FOOTBALLDUDE88]

I think that the playability of Marvel depends on which one(s) you plan on using. We play 500 points and occationaly use Marvel and we can usually get them to fit in. One time I had Thanos and he got killed fairly early on in the game. My two opponents killed every other figure in my army exept one venoc viper and they didn't think that they needed to worry about it and I roled a 20 for rejected by death and managed to kill each of their Marvel heros and won the game. That is my Marvel story.
SO IMO GO THANOS!!!!!!!!!!!!!!!!!!!!!!!!

[IAmBatman]

Hey, now that Battle Royale is over, it might be worth it to change that really annoying signature, FootballDude.

[Eirikr]

I don't think it is points that greatly effect the usefulness of high cost Marvel heroes. Rather it's an issue of points/startzone-spaces. I have found that about 2 startzone spaces for every 100 points makes for a very fun mixed marvel game where big heroes are very nearly a must in every army. I have a rather long list of figures, about a quarter of the classic figures, divided into 2 factions and the Marvel heroes split into heroes and villains. It makes for a really rich experience but is definately in the Marvel world, not Valhalla. I will find my lists and post them to give you an idea of what I mean.

[FOOTBALLDUDE88]

Hey, now that Battle Royale is over, it might be worth it to change that really annoying signature, FootballDude.

Happy????????

[IAmBatman]

Very much, thank you. :)

[FOOTBALLDUDE88]

Very much, thank you. :)


You are very welcome.:)

[Eirikr]

Very much, thank you. :)


You are very welcome.:)
Your are all so friendly and polite. Thank you.:D