![]() Of those 220 unique decks, 41 had enough representatives to make the tier list. This is true of a number of other decks too, but there's not much to be done about it this go-round. And those are the most similar decks I called almost every Grixis-colored deck without Shadow "control." Therefore, its numbers are arguably inflated. Recognizable as derivative, but not exactly the same deck. These days it's more like Death's Shadow-less Grixis Shadow. In March, that deck was a pure control deck. Remember my opening confession? That applies here. Between March and December 2021, 4,811 individual results were posted, representing 220 unique decks. Unsurprisingly, with 10 months' worth of data included, the number of decks being analyzed is staggering. Which sounds very high, but actually isn't in aggregate. Subsequently, to make Tier 1, 128 decks are required. Tier 2 starts with 75 results and runs to 127. That's an STdev to make statistics projects worry, but it makes sense given the data set. Again, it's the starting point to the cutoff, then one above for the next Tier. The STdev was 51.77, which means that means Tier 3 runs to 74 results. Then we go one standard deviation above average to set the limit of Tier 3 and cutoff for Tier 2. Tier 3 begins with decks posting 22 results. (I told you this would like the monthly updates, just with a twist.) ![]() Being a tiered deck requires being better than “good enough ” for 2021 the average population was 21.96 setting the Tier 3 cutoff at 22 decks. The average is my estimate for how many results a given deck “should” produce on MTGO since that's what's being sampled. To make the yearly tier list, a given deck has to beat the overall average population for that year. To any statisticians out there: I'd love to hear what you would have done! 2021 Metagame I've decided to ignore this and press ahead because it won't affect the conclusions. By most normal tests, every Tiered deck is an outlier. In short, the number of decks making the tier list is high, but at the same time, every outlier test essentially fails. The standard deviation (STDev) is extremely high thanks partially to time pressures, but also to the vast distance from the bottom to the top. It also means that there are a ton of singleton decks, which significantly alters the data. That means that decks that have disappeared from the current metagame are not only present but have a strong impact on the data. I did not include the data for January because it represented a metagame that was banned away, and I didn't bother doing an update for February for the same reason. Specifically, a normal update is for one month, while this one comprises ten. What is very different is the scope of this analysis. Mechanically, everything is the same as it is for a normal monthly metagame update. I literally just added each month's total data together to generate the total data for the year. It is just like any other metagame article I've done over the past 18 months because all I'm doing is aggregating the data I've previously reported. This both is and is not a typical metagame article. I'll be better about watching for that next time. This created considerable headaches once I combined the data because I reused a lot of names for decks that changed considerably over the year. If a deck didn't consistently show up on the Tier list, there wasn't much need to worry about what it was called. Because I didn't plan this out, I didn't worry about full deck-name consistency month to month. Thankfully, comments coming in around August put it back on my radar. It didn't occur to me to combine the data for the full year last year. I feel the need to lead with a confession. Today then, I bring you the full metagame for post-ban 2021 Modern. Which is my way of saying that I have an entire year's worth of data accumulated, and now it's time to display and explain it. Sorry to disappoint anyone, but I can't fully close the books on last year until I have fully unloaded the books about the year.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |