Saturday, June 7, 2014

The High Cost of Dice 2: Dice Harder

As we've mentioned several times on the podcast and in articles here and there, we at The Reserve Pool kind of get the feeling that there are too many expensive characters in the game - that there's too many large, smashy guys, and not enough options for a smaller, agile game.
Given the relative difficulty in fielding said big guys, and the importance of getting early damage in so that capitalizing on a later opportunity is more likely to result in a win, this is probably part of why we see the same few characters coming up over and over again when we talk good characters for decks - the best of our limited lower cost options, like Tsarina and Beast.

That got me to wondering - just how bad is it?  I decided to sit down with Dice Masters DB and Clear and Draw, plus a handy Google spreadsheet, and find out.

Guess what I learned?  Spoilers: dice are expensive.

The Purchase Cost Is Too Damn High

The first thing I wanted to look at was the average purchase cost of a die.  This was simple - I took the average of each character's versions' purchase costs, and took the average of that.  Including Mjolnir and Vibranium Shield, we wind up with an average cost of about 4.83; without them, 4.81 - in short, if we had a team of 'average' cost cards, we'd expect it to be mostly 5s, with a couple of 4s tossed in.  I don't see a lot of squads running even vaguely that high, usually having maybe one 6 or 7, a couple of 5s, some 4s, and a couple of 2s.

Why this is the case is clear when we take a look at a graph of the non-action card costs.

There are too many 6-cost cards!  We've got way too many of those compared to 3s, which they should be 'balancing' - even if we're assuming a more bell-shaped distribution (in which case the 2s and 7s are fine: 12 of each), it's totally out of whack, and skewing everything upwards.  If you remember what it takes to actually field a 6-cost character - at best, two characters or actions and two sidekicks, fielding nothing that turn, and likely more like eight dice if you want to bolster your defenses - it seems very out of place to have so many of what are seem to me like capstone characters.

It almost seems like they did in fact want us to have two or three in our squad, and plan on getting them out. However, given how much that I feel that would slow down the game, it doesn't make any sense in the context of 15-minute tournament rounds.  That's a pretty short span to go through what would probably wind up including an extra five or six turns of jockeying for a time when you can spare enough dice from your defense to get that magic six energy.  Speed Dice Masters, I suppose.

Because I was curious, I also took a look at average purchase cost as a function of energy type; there's actually not too much variation there, with Shield coming in a little low at 4.76, Bolt and Mask at almost average at 4.79 and 4.81 respectively, and Fist fairly high at 4.98.  Fist came out high because outside of Widow, it has nothing to counteract the high cost of its Big Smashy Guys like Hulk and Thing.  By the by - the energy types are roughly even in count, with Shield/Fist having 9 and Bolt/Mask having 10.

The Fielding Costs Are Too Damn High

The average TFC of the Avengers vs. X-Men set is 4.08.  Given that, the 'average' character would be expected to have roughly a 1-1-2 cost structure.  But what's the actual distribution look like?

Well, the 1-1-2 crowd is the second-largest group...but the largest, by a 50% margin, is the 1-2-3 crowd.  This makes even less sense given all the 6-cost characters; how the heck are we supposed to field all these mega-expensive characters and still scrape up a bazillion energy to buy more of them?

Considering you need to use at least two dice to put a 3-cost character out, it feels more and more like we're expected to be rolling eight or more dice frequently to pay for all this stuff, which to me would imply a firm base of lower cost characters for us to build on, both in purchase and fielding cost.  This makes sense, and we do have some base there, but the number of options pale in comparison to the choices we have at the high end of the spectrum; we end up going for the same few cards or dice for that meat and potatoes we need during the early-to-mid game, and all the decks get all samey.

In terms of average total fielding cost as a function of energy type, Fist makes up for its poor showing in purchase cost by being significantly lower than average (and the other types), with a 3.78.  Shields are extra expensive to put out at 4.38 on average, with Bolts being slightly cheap at an even 4, and Masks a little expensive at 4.2.  Fist has several very cheap characters (Widow, Deadpool, and Spider-Man) helping keep its bottom line down, while Shield has no TFC 1 characters, and only 2 2s, with the rest being 4 or more, resulting in a higher average.

The Stats Don't Favor The Characters With Costs That Are Too Damn High

The last thing I looked at was stat distributions.  Average stat values are refreshingly straightforward* in relation to fielding cost.
Stats increase with fielding cost, to no one's surprise, with about 1.5 attack gained per point of fielding cost, and about 2 defense per point.  Accordingly, defense usually trends higher than offense, and I'd noticed this before putting together the spreadsheet.  The dice seem to be a little more generous with extra defense compared to extra offense, with more high-defense characters like Deadpool, Mr. Fantastic, and Loki than high-skewed offense characters like Hawkeye or Punisher.

Looking at average stat values against TFC, though...

...we see a few things.  First off, the 3 TFC guys are pretty wonky - they include Mystique, who throws off the curve because her (terrible) listed stats aren't usually what she's going to be using.  More notably - the TFC 5 characters aren't really an improvement over the TFC 4 characters statwise, and while TFC 6 characters gain a good point in each stat over those, the average TFC 6 statline is barely equal to the sum of that of two TFC 2 characters.

Considering the relative lack of trample effects in the game presently, I'd argue that this again suggests you're better off using those smaller characters - they're safe to block big guys (and give you churn doing so), give you more opportunities to get through, and can deal roughly the same amount of damage for the same cost.  You do run into some throughput issues - you can only draw so many dice per turn, so you can get more potential damage in a single turn out of big dice than you will small.  That's offset by the smaller characters generally being easier to force to churn; additionally, how often can you manage to draw two or three of those big guys and pay for them in the same turn?  I wouldn't say they don't have their place, but the stats just don't seem to back up how often their frequency suggests we should use them.

The Conclusion Is: Too Damn High

So what we've learned - had confirmed, really - is that there are disproportionally many higher-cost characters, and relatively few really low-cost ones.  If you think about it, the field is even worse that you'd think for the 2-cost characters; half of them are on two characters (Black Widow and Beast), one of them is that stinker Johnny Blaze, the Storm and Human Torch options are significantly inferior to some of the other choices for that die, and Nick Fury is pretty situational.  That's really not a lot to choose from, especially in contrast to the diverse choice of TFC 6 brutes that end up as 8/8 we have.

What can we do about this?  Errata to change the set this drastically seems unlikely, and while there are a few new cards coming out as OP prizes, due to their limited availability, I wouldn't expect them to be agents of change in this vector, either (plus, most of them seem to be for the beefier characters, anyway).  It's possible that the new basic actions could breathe some life into the big guy roster, but the two we know either don't particularly help (Rally) or actively benefit a weenie-style squad (Teamwork).  So, overall, our hopes will probably need to be pinned on Uncanny X-Men.

Join me next time, when I'll take a look at the one of the most important kind of cards: those that help us draw more dice!

*: I'd also made a version of this chart that includes the level of the character, but as you can guess a three-axis comparison is sort of hard to visualize in a two-dimensional medium.  It's here if someone wants to look at it, and shows largely the same information as the chart that doesn't split out by level; the only real point of note is that there's a drop in stats for 2-cost characters going from level 2 to 3.  This is because the number of level 2 characters with 2 fielding cost have their stats inflated by all the TFC 6 characters progressing to their (usually 8/8) level 3 face, and those all drop out for level 3.


  1. Nice analysis and it's illuminating to look at how costs break down, but I fundamentally disagree with the premise of this article. There are NOT too many high cost characters - on the contrary there are too many ultra powerful el cheapo characters that make it more or less impossible to play them. The large number of 6 and 7 characters should tell us something about how many of those guys are intended to be used in a given game. Answer: a lot. In addition, there's been some disconnect on the length of games - most have thought (you in particular, Evan) that the Wizkids estimate of 15 mins is too short. Well if there are a bunch of heavy hitters out there games end pretty quick and usually in a blaze of glory. So two separate things, a) sheer numbers of characters and b) the intended length of games point me in the direction that the designers were intending that the bulk of a team and the bulk of the strategy rested with high cost characters and how they battled against each other. Instead what we have is teams entirely built around Black Widow, Beast, Gobby and Human Torch. Because they are cheap to buy, cheap to field and absolutely devastate the opponent. In my humble opinion these cards were simply not playtested enough and they should be costed higher. The low cost guys should all be fairly crappy characters that act as a ramp to the big guys while offering a clear advantage over an opponent who thinks they might want to go with just sidekicks and try to jump right to the heavy hitters. But unfortunately this is not the case. The low cost characters like Beast, Black Widow and Green Goblin are very clearly some of the best cards in the game which should simply not be the case (Human Torch: Johnny Storm is a special case in my view - he's not super cheap at 4, but should probably be a 6). I have a house rule that Tsarina is a 3 cost, Gobby is 4, Human Torch: JS is 6 and Beast: 666 is a 4 cost. Try a few games out with those costs and you will find the game opens up quite a bit. I like when Phoenix and Thor and Colossus and Dr. Octopus are out there mixing it up - playing the 100th game of "Tsarina and the Gobblets" is getting a bit old.

    1. Whoa, sorry for the late reply. This is absolutely another good way to look at it - that they screwed up on making some of the low-cost characters way too strong, and really did intend for longer, more high-powered slugfests. Overall? I'd agree and probably say that "bigger" fights were their intent, and that the 15-minute rounds was just a bad guideline on their part, but I'm not sure that making a lot of the better cheap cards would help the game so much as just make it slower. I'd have to give the revised costs you propose a shot and see what happens, but honestly I think I'd be happier seeing cheaper, more risk vs reward-ish power cards rather than piles and piles of 6-1/2/3 characters and turns doing the high-energy-when-I-don't-need-characters-too-bad raindance. Speed isn't bad if the game was balanced around it, but they seem to have goofed with that here.

    2. The only part I'd disagree with here is the idea that high powered slugfests last a long time, since I think it's clear that once you get a bunch of bruisers out, you're in the endgame. But the game should build to that - right now it's just field/distract/force attack/distract with the same low cost characters with ping damage being the norm. It's counterintuitive, but to me, THAT is the version of the game that actually lasts longer. Overall it's just way more logical to think that they screwed up on a handful of low-cost cards rather than they completely misread the gameflow and purposely added 40+ highcost cards that you'd never be able to use.

      By the way, just listened to the latest podcast and I think your idea of "weirder" dice and characters is a cool idea. I'm talking about dice with unbalanced, non-linear fielding costs and also characters with a built in "penalty" for fielding/having active. Hopefully that is part of the Uncanny set.

  2. Also, you ask in your article where you'd expect to get the energy to field big 6-TFC guys with such a lack of small characters. I would argue that's where the cheap, energy-rich action dice come into play (another barely-touched part of this game that was clearly intended to loom large). It's nearly impossible for me to assume that the designers actual intention was for large parts of the game to be irrelevant or useless. The only way I can see this changing is cost errata as described above or perhaps more stringent dice limits on the cards I mention.

    1. This comment has been removed by the author.

  3. I don't think we'll see an errata either, and with the sheer volume of new cards and characters and new mechanics planned to come out I have a feeling any new sets will be even less playtested than this one seems to be.

    In the end, I'm going to play it my way as it is far more fun. In fact, most of the real problems with this game (like actions being suboptimal, high-cost characters being inaccessible) go away when you do as I describe and, to that end, I really suggest you try it. With a 3 cost Tsarina, there is actually a reason to use the other Black Widow cards!

    In terms of going forward, I actually think there will eventually just be banned cards just like there are in Magic. Rather than go the lazy approach (which I consider outright banning to be), it's a lot more fun to just adjust the cards. This game is FAR more popular than Heroclix has ever been, so not sure we can use that game as a guide.

    In any case, I'm not willing to just shrug my shoulders and write off close to 60% of the characters in this game just because of sloppiness on WizKids part. I'd much rather try to make it work the way it was clearly meant to.

    1. Changing it on your own is perfectly fine except that's not going to work in OP.

      As far as HeroClix goes, I'm talking about how WizKids has shown themselves to operate for years. Set comes out, six months of story OP, then they never touch that set again. This has been their history as a company. I'd like that to change, but honestly if this game continues to be THIS popular and sell so well that they can't keep it in stock, there's probably more motivation to come out with sets than there is to fix old ones.

      Even so, I was talking with a guy on Twitter who said - and I agree to an extent - that trying to win with Tsarina and Gobby is like trying to win at Chess with just your pawns and so forth. Try out that control build, or it's X-Men-centric cousin. There are two decks that have nothing to do with these two meta defining cards and can win in their own right.

      There are always some useless cards in these games (even in the LCGs, where the vast majority of every Netrunner data pack is chaff), and I wouldn't say that it's anywhere near 60% for MDM. If I had to go through and single out "cards I'll never use" it would be a pretty short list.

  4. Lol that was me - I am blitzing social media! I don't mind some cards being "better" than others, but this is different - it's not about cards being "good" or not - it's an entire class of cards that is virtually inaccessible due to how the gameplay flows. They may be awesome cards, but you'd never know because they're too expensive. The current meta is too fast and cheap and doesn't allow for that. I'm sorry, but I just don't believe that's how this game was supposed to go.

    You use Netrunner as an example - I see where you're going with this but I don't think it's the right comparison. Sure there are certain cards that are good and bad, but there is no type of card or cost-point of card that is unusable en masse. Big expensive ICE can be used to great effect in Netrunner!

    Regardless, the entire point of this post was to say, "hmm, why are there so many high cost cards I can't use?" and all I am saying is if you make a few tiny changes you have a world where you actually can use them.

    1. Here's the thing, though - they definitely intended speed, else organized play wouldn't expect three games in 45 minutes.