Exactly what was reported. All the individual results are out of 50. So for the Affinity matchup both the test Jund deck and the control Jund deck won 28 matches and 28/50=56%.

Tron may or may not be able to hold down Jund, but it isn’t the only big mana deck out there. For example, I doubt that Valakut decks suddenly become good matchups because of Bloodbraid.

So what are the actual win percentages for each then? These percentages obviously doesn’t make it very clear to us novice statisticians because, to me, the “total wins” would be the decks actual “win percentage”? Obviously, this is not the case, so, I’m unsure of how to get the actual win percentage from these results.

Makes sense. It’s just that I’ve always viewed Jund vs. Tron as the defining polarizing matchup of the format, the epitome of a matchup you might as well give up since there’s nothing you can reasonably do to improve it. So if BBE actually improves Jund to an even matchup, to me that’s ringing some serious alarm bells. If big mana can’t prey on Jund, what can?

You are correct that each of those improvements are marginal, though you put enough marginal buffs together and they add up. The real reason is how the matchup dynamic has changed. Jund wins by doing something slightly more powerful than you at every point of its curve. Tron wins by doing something significantly better on turn three and each turn thereafter. That’s why Tron’s such a bad matchup, it’s overpowering everything Jund has and requires either specific answers and lots of them. The way to beat Tron is by swarming it. Creature swarms or decks that can just play lots of spells make up for that individual power deficit in bulk so that Tron struggles to keep up (unless they have Ugin). Bloodbraid lets you play two spells a turn and that quantity helps with the power deficit, which greatly improves the matchup. I don’t think it becomes favorable, but even is definitely possible. Remember, Tron preyed on Jund in 2012 but it wasn’t oppressing it.

The Ponza Elf decks were usually Zoo-style and ran Boom//Bust (do a search on mtgtop8), but that isn’t a hittable target anymore with the CMC rule change on split cards. Not saying Elf wouldn’t improve the RUG wedge generally, just that it wouldn’t suddenly birth a good RUG deck, as it doesn’t solve any of the combination’s problems.

Cascade doesn’t care about what is on top…you keep flipping til you find something to cast.

Tell you what though, as an extra option in the card pool, RUG players would rejoice at having the option, even if jund overall was a better fit for the card.

Some sort of aggro/ponza list? I can recall these decks existing back around kamigawa standard and I’m sure stone rain was hit off cascade many times in modern before the elf was confined to a cage :D. Not proposing any sort of top tier viability, but most of modern is brewtopia anyway. The tiered, visible portion of modern is such a small (albeit influential) slice i’m sure RUG would be happy to welcome the elf back to the fold.

Regards,

RUG lacks decent hits besides AV. Goyf and Scooze as threats, sure, but then what? Bob, Lili, KC, Decay, Brutality… black just has so many amazing cascade targets. And RUG can’t really interact with noncreature spells because permission clashes with cascade and blue lacks targeted discard (not that your creature removal is even passable in these colors). I’m pessimistic.

You are reading it wrong. When I did the individual stats I said Control/test deck wins: total wins, win percentage. You are reading the total wins as the win percentage.

I do like that you’re splitting the subjective from the objective more clearly with the staggered articles. And I think it’s fair to just say with no adaptation by the meta bbe improves a number of key matches for Jund by around ten percent. I’d have to go back and see how that number compares to your other tests. I feel like it is way higher than the stone forge results.

I also think it’s tough in the abstract to know what percent increase would be dangerous for the format. Its relative obviously as moving a twenty percent deck up to fourty is fine but moving a fourty five to sixty five is not. So there’s probably an absolute you don’t want to cross like sixty percent vs the gauntlet?

This is what next week’s article will be addressing.

Incredible effort by the way. All in the name of learning more, it’s really great to see.

Let’s examine a couple of things you’ve said though:

1) tron matchups. You indicate that more wurmcoils would almost invalidate the added bonus from including bloodbraid. I’m inclined to agree. What’s tricky here is what this means in the wider context. Tron players often fluidly shift around on numbers of cards depending on the metagame. You can bet that within a week or so, if bloodbraid was unleashed, the composition of threats would change accordingly, thereby making your results here a bit less valid.

What’s more, every deck would do this, to varying degrees. It’s possible that your snapshot from which you are taking the sample matches may be giving unfairly weighted results :S

Although on the other hand, modern (when taken as a whole) is a vast soup of decks and players often deviate from a generally accepted ‘optimal’ build, so the counter argument is that at least this data means something in the abstract, even if it doesn’t apply to a changing fluid metagame.

I don’t envy your defense of these data. If this was a postgrad thesis i’m sure the prof would have a lot to say 😛

Either way it was an enjoyable read and I’m thankful for your effort here. I don’t currently know of anyone else doing something like this so it’s not only a novelty but there might be some interesting lessons from it as well.

Like I said, it is weird. I definitely saw Santa a lot this time, I’m not denying it. That’s why I never liked Bloodbraid back in the day, it always seemed like that was the norm. Though it was three Fulminators not two.

As for Burn, it wasn’t that bad a matchup in 2012 and I imagine it wouldn’t be the worst now either. Yes, Bob is a huge liability there but there have been times when Jund ran him alongside Hit // Run, so maybe you don’t care. You do still board him out against Burn, an Thoughtseize actually isn’t bad against Burn since it’s functionally gaining one life per burn spell it takes. Inquisition is miles better though.

I do still think finding land or elf on top two thirds of the time isn’t right when those cards are more like a generous fourty percent of the deck than sixty six.

Similarly your board had two finks two fulminator and an anger. I don’t see many matches where all five come in, and on game one you have eleven one cmc cards vs six three cmcs anyways. The best scenario would maybe see your count of one’s and threes balanced putting avg cascade cmc squarely at two.

Basically a ten percent jump in wins seems pretty high and I wonder if you just incidentally ran into an awful lot of Christmas lands with your cascades. I would also expect you lose serious points vs burn or other fast aggro with a set of seize and bobs and bobs flipping elves.

About 1/3 of the time there was a target on top. The mode of the cascade data is two by quite a large margin, however I had enough long cascades to pull the average up significantly, including 16 instances of cascade 6.
And yes, the 2.7 is weird to the point of anomaly. I did hit Liliana and Kommand an inordinate amount of time, but the data set also included sideboard games where I had far more 3-drops and that shifted the average up. Also, between three draw steps minimum and often some Bob triggers, you tend to draw a lot of your 1- and 2- drops naturally before you could play Bloodbraid at all, making it more likely to hit high drops.

I personally would like to see Bloodbraid Elf come off the banlist. Modern is a format of “unfair” decks and bloodbraid is the exact sort of card non-blue decks need to keep up with the nonsense in the format.