Happy Election Day, America! May your vote be counted and the voting process as stress-free as mine. Living in a state that’s been all mail-in ballots for years has considerable advantages in a pandemic year. And if not, take heart. At minimum, you won’t be bombarded by annoying, omnipresent, and pleading political advertising for at least another two years.
October’s data came in like clockwork. Not only exactly on time, but also in predictable quantity. Every week there were exactly two Challenges and five Preliminaries. This appears to be the standard for Wizards and what I’ll expect going forward. If there are additional premier events, that’s just gravy. I’ve considered adding in non-Wizards MTGO events. MTGGoldfish and MTGTop8 often have other events from international groups and Hareruya, but there are problems. Primarily, I don’t know the competitiveness of the events. For all I know, they could be small playgroups with really inbred metagames, not at all comparable to MTGO events. Also, a lot of them report weird quantities of decks, like Top 11’s or Top 4’s, or 27 total decks. I’m not sure what to do score-wise, so I’m not going to bother. Let’s worry about that when paper starts again.
There were 585 total decks this month. Which is down a bit from September, but that’s largely thanks to no Modern Championship events this month, meaning fewer Premier events to draw from. Preliminary attendance appeared to be up overall to compensate for the lower event total, and so the data is only down by 26 decks.
To make the tier list, a given deck has to beat the overall average population for the month. The average is my estimate for how many results a given deck “should” produce on MTGO. To be a tiered deck requires being better than “good enough”. For October, the average population was 8.47, meaning that a deck needed 9 results to beat the average and make Tier 3. Then, we go one standard deviation above average to set the limit of Tier 3 and cutoff to Tier 2. The STdev was 12.71, so that means add 13 results and Tier 2 starts with 22 results and runs to 34. Thus, to make Tier 1, 35 decks are required.
The Tier List
Deck diversity is down a bit this month, from 73 to 69, which is still a nice amount of variety. Of those, 21 cleared the threshold and made the tier list. The lower numbers are almost certainly a function of the smaller data set, which was caused by there being fewer premier events in October. However, population differences cannot explain how skewed the data is.
|Deck Name||Total #||Total %|
|Death and Taxes||35||5.98|
|Oops, All Spells||18||3.08|
|Niv 2 Light||9||1.54|
4-Color Omnath piles and Scourge Shadow technically qualified for Tier 0 status (three STdev’s above average) in October. That is an absurdly high population. Ponza was just under the cut last month. I’m not going to actually call either deck Tier 0, temptingly click-baity as it may be. Both decks lack the decisive bustedness of a Hogaak or Eye of Ugin Eldrazi, and it’s also only the first month this has happened. This could easily be, and I suspect it actually is, a function of variance and popularity rather than a problem.
See, 4-Color Omanth covers a huge variety of decks. Some are Bant Uro decks with a single Omnath, Locus of Creation thrown in. Many are completely retooled to be Omnath first, planeswalkers second, Uro third decks. And everything in between. If I wanted to, I could have broken Omnath into at least three decks. I’m not going to because I hate splitting hairs. But I could have, so don’t freak out over the numbers. Meanwhile, Rakdos Prowess has been a popular deck every month so far, and its evolution Scourge Shadow got a lot of attention in October. That boosted its numbers a lot. As the rest of the data showed, this is nothing to worry about.
Where’s Tier 2?
There’s only one deck in Tier 2: Reclaimer Titan. That’s just how the math worked out. For those that don’t know, it’s basically Amulet-less Titan using Elvish Reclaimer and Flagstones of Trokair for mana fixing, toolboxing, and value. It’s the evolution of the toolbox Vial Titan decks that showed up in August, then disappeared. I’ve never played against them and so have no further commentary, but they did better than typical Amulet Titan, so there’s that.
As for its loneliness, I think that’s just a quirk. With so much population wrapped up in the top two decks, it’s natural and expected that the rest of the standings would be a bit thin. Death and Taxes is literally just over the line for Tier 1, and a number of decks are just under getting out of Tier 3. I don’t think there’s anything to be worried about metagame-wise as a result.
It does bring up some methodology questions. In the old system, this would never have happened, thanks to the paper data outweighing the MTGO data coupled with the statistical “smoothing” of the weighting system. I strongly believe that 4-Color Copycat, Izzet Prowess, Heliod Company, and Amulet Titan are actually Tier 2 decks in October’s metagame, but the high STdev kept them out. I’m not sure there’s an acceptable solution to this, as without another data source to serve as a counterweight any smoothing I do would be illegitimate. Weird-looking Tiers may just be something to live with until paper resumes.
What Happened to Ponza?
Meanwhile, I don’t think I’ve seen a crash quite like Ponza’s. It went from the best-performing deck in Modern to just barely making Tier 3. Insert tasteless airline joke here. Honestly, I’m perplexed. The metagame overall doesn’t look too different from what it was in previous months, and given Omnath’s percentage, you’d think that a Blood Moon deck would be ready to pounce. Instead, the deck died. I understand that Ponza has a poor Scourge Shadow matchup, but is it so bad to completely eliminate the deck?
Tron also crashed, but that is far more understandable. Eldrazi Tron has always been a meta deck, specifically a Chalice of the Void meta deck. When Chalice loses power, like when prowess continues to decline (as happened in October) Eldrazi Tron dies off. Mono-Green Tron also collapsed, but this was foreseeable. Cleansing Wildfire is a tailor-made maindeckable anti-Tron card. Not that it’s being maindecked as an anti-Tron card, mind you. There are a lot of decks throwing Wildfire at Flagstones of Trokair to make their own, better version of Rampant Growth. They’re just incidentally getting to use it for its intended purpose.
Tracking the metagame in terms of population is standard practice. However, how do results actually factor in? Better decks should also have better results. In an effort to measure this, I’ve started using a power ranking system in addition to the prevalence list.
A reminder of how it works: as I go through the Preliminary and Challenge results, I mark each deck’s record or placement respectively. Points are then awarded based on those results. Preliminaries report results based on record, so that’s how the points are distributed. 5-0 is three points, 4-1 is two, and 3-2 is one. Challenges are reported in terms of placing, so being Top 8 is worth three points, Top 16 is two, and being reported at all is one. The system is thus weighted to award more points to decks that perform well in Challenges rather than Preliminaries. The reason is simply that Challenges are larger and more competitive events, and the harder the field, the better a deck needs to be.
The Power Tier
The 585 decks earned a total of 918 points in September. Total decks were down, so it makes sense that points are also down. The average points were 13.30, so 14 points makes Tier 3. The STdev was 20.71, meaning Tier 2 began at 35 points and Tier 1 is for 56 points or more. There were 20 decks in the power tier, down Ponza from the population list. It was not only abandoned, it performed poorly.
Again, the data shows much higher variance than before. Given that the data is linked, it would be far weirder if that wasn’t the case. Interestingly, the difference between the average and STdev for points has consistently been double that of the prevalence rankings. That the STdev would be higher makes sense, there’s more variation options for ranking power than population. That the relationship has been so consistent is intriguing.
|Deck Name||Total Points||Total %|
|Death and Taxes||56||6.10|
|Oops, All Spells||28||3.05|
|Niv 2 Light||15||1.63|
The top performing decks earned an absurd amount of points this week. To be expected, perhaps, given the population data. However, the Omnath decks really took the cake. They won a lot of events, but that may not be cause for concern. Pilots matter, and if the best players all gravitate to the same deck it will always do better than anything else. I can’t prove this is the cause, but it is a plausible explanation. Don’t go losing your mind until more data comes in.
The advantage of doing power rankings is that some of the population weirdness disappears on its own. Tier 2 has naturally expanded to three decks, though not the ones I expected. Jund shot up from Tier 3 while Heliod, Izzet, and Amulet languished. At very least, it confirms the relative power indicated by the population listings. As does Eldrazi Tron being the lowest scoring deck in the listings.
Average Power Ranking
Finally, we come to the average power rankings. These are found by taking total points earned and dividing it by total decks, which measures points per deck. I use this to measure strength vs. popularity. Measuring deck strength is hard. Using the power rankings certainly helps, and serves to show how justified a deck’s popularity is. However, more popular decks will still necessarily earn a lot of points. This is where the averaging comes in. Decks that earn a lot of points because they get a lot of results will do worse than decks that win more events, indicating which deck actually performs better. A higher average indicates lots of high finishes where low averages result from mediocre performances and high population. Lower tier decks typically do very well here, likely due to their pilots being enthusiasts. So be careful about reading too much into the results.
The Real Story
I’m still searching for a more statistical way to quantify how “good” decks are. All the tools from sports statistics deal with individual players on a team, which might work for individual cards, but not deck-to-deck. If anyone has ideas, I’m open to suggestions. Until I find something better, the averaging system is still very instructive. The average average was 1.61 this month. Therefore, any deck that is above that baseline average overperformed while those under it underperformed. Interestingly, 1.6 with some change has been the baseline for every month I’ve done this so far. No idea what it means, but it is an interesting discovery.
|Deck Name||Average Points||Power Tier Placement|
|Niv 2 Light||1.67||17|
|Death and Taxes||1.60||4|
|Oops, All Spells||1.56||11|
There really is no substitute for experience. Jund was the best-performing deck by a solid margin. Supposedly dead thanks to the Uro/Omnath decks, Jund players defied October’s long odds to pull a solid 2.00 point average. Jund’s gameplan is just too solid and adaptable to really keep down, and it will be interesting to see how November goes for the stalwart.
Developments to Watch
Uro decks account for 4/9 of the overperforming decks, and that’s concerning. A popular deck being popular is one thing, but if it wins a disproportionate amount then we’re getting into a ban watch. There’s enough variation between individual decks that diversity isn’t being affected, both between the decks and inside them. It’s also weird that it’s working so well, given my experience with these decks. It makes me wonder if there’s an actual problem in the meta or there’s groupthink and self-fulfilling prophecies going on.
Meanwhile, this month’s runner up Scourge Shadow decidedly underperformed. The deck is solid, but it very clearly only made Tier 1 due to popularity. This indicates some combination of high variance, poor metagame positioning, and a flawed attack plan. Changes need to be made. Finally, and mostly because I like crowing when a deck I think is bad does poorly, not only was Eldrazi Tron on the bottom of the power rankings, it had the worst average. The deck is badly positioned and clunky now that prowess has fallen off and Chalice of the Void‘s poorly positioned. Time to put the spaghetti monsters down.
Keep on Keeping on
The metagame has begun stabilizing, which is appropriate as we move into the period that is traditionally the least dynamic for Magic. Competitive play has always wound down November to January in the past. This time will be a bit different I suspect, as the draw-down was for paper, not MTGO. I expect the metagame to just keep plugging along in November, but we’ll see what happens.
David began playing Magic during Odyssey block, quit playing Magic when Caw Blade ruled the world, and returned to Modern shortly before Deathrite was banned. He’s made an appearance at the Pro Tour, made money at GP Denver, and is constantly grinding and brewing in Modern.