nvm that they dont show the lists or players or matches that vary quite a bit, and that its a hugely luck based game, and that one single match gone the other way swings it 2 whole percentage points.
Which is exactly the reason why there is statistical hypothesis testing there. Like, literally, that's the point.
statical evidence that it's good...FOUR of the top eight decks had FOUR of them....that's real statistics. And stopping a sword plan game one isn't really all that needs to be done. Gladiator can be ousted and does absolutely nothing...and when you DO attack with it? its just a 4/4 and thats if they DONT tumble magnet it. Also...inferno titan unblocked does incredible damage.
That isn't a statistic. You aren't comparing it to anything.
I would like to be able to see the 4-0 and 3-1 decks from MTGO daily sealed events, like they currently have for the constructed formats. However, at the present time, Wizards does not post them. There is a petition of sorts on the MTGO forums. You can find it here. If you would be interested in see this information as well, make a post there. With enough demand, Wizards will do this for the community.
Inferno Titan shows up more frequently in 3-1 decks than 4-0 decks on Magic Online. Cyclops Gladiator, on the other, shows up far more frequently in the 4-0 decks.
I really think UWb is the better deck, but others might not agree because of their own experiences, and they are entitled to that, but don't make up percentages that are not correct and call them fact, just cause at your FNM you 3-0-1 every week doesn't mean your deck has 75% win against the format.
Don't be a mindless drone to the statistics someone else collected and manipulated for you. Play both decks. Playtest both decks. See what works for you.
However much playtesting you can do will not be nearly as informative as the large-n data. I am not claiming you should not playtest. You should. But the goal of playtesting should be to understand how to play your deck effectively, not a question of "is deck A better than deck B?"
It's definitely your opinion, all considering when you say "WE REMOVE GERRY AND LOOK THE STATS GO DOWN" Because you don't have the stats, nor the proper reasoning to do so. Which is just statistical sin. There is no excuse, this issue should have never arose, nor is it factually validated. It's just opinion, I have no patience for that.
Although it would be perfectly fine to do this anyway, Gerry Thompson is not removed from the statistics presented.
I strongly recommend you discount anything coming from Spaniel's statistics. The algorythms he uses slants the actual results. MTGO's metagame is also absolutely nothing like paper magic in regards to the percentage of each being played.
The results are clearly robust to any changes between the online and real life metagame. Look at the matchup data.
And this kid is the mtgo kinda guy. His "statistics" on the subject say otherwise. Now I personally haven't done any actual math on the subject. But there are things I do know, 1, is that UWb has been out as a deck for less time and therefore has to "make up" for lost time which skews success ratings, and, 2, looking at the top 16 of the last bunch of SCG tournaments. even without GT the UWb deck has had more top 16 than the UW one. Whether you do math or not, showing up in the top 16 MORE OFTEN in recent tournaments than the other deck surely means the success rating of the deck is most likely higher. If you are counting tournaments that UWb didn't exist in, yea the success of UW is higher, because the other deck didnt exist yet.
No, it means that UWb is played more frequently. Reid Duke's data have a laughably small-n. Every article he presents basically tells us nothing about the metagame.
yea caw-blade came out months ago and many many tournaments ago, so yea, you are going to have more top 16's than a deck that came out weeks ago... So congrats on your deck being in more tournaments I guess But 60% against the format is incorrect.
This has nothing to do with the comparison of UWb to UW presented.
Which is exactly the reason why there is statistical hypothesis testing there. Like, literally, that's the point.
That isn't a statistic. You aren't comparing it to anything.
Nothing does particularly well consistently. Maybe RDW and GW Quest. I guess you could throw Eldrazi in as a possibility.
The statistics are on Cyclops Gladiators' side. Maindeck too, not sideboard.
Inferno Titan shows up more frequently in 3-1 decks than 4-0 decks on Magic Online. Cyclops Gladiator, on the other, shows up far more frequently in the 4-0 decks.
Let the debate rage.
???
This is not the case. It's just about exactly as much played last week as it is now.
http://magic.tcgplayer.com/db/article.asp?ID=9569
Whether it stays that way remains to be seen.
What do you think the author of Power Rankings (me) does for a living?
http://wjspaniel.wordpress.com/
gametheory101.com
It is.
Who is making up percentages?
However much playtesting you can do will not be nearly as informative as the large-n data. I am not claiming you should not playtest. You should. But the goal of playtesting should be to understand how to play your deck effectively, not a question of "is deck A better than deck B?"
Although it would be perfectly fine to do this anyway, Gerry Thompson is not removed from the statistics presented.
The results are clearly robust to any changes between the online and real life metagame. Look at the matchup data.
No, it means that UWb is played more frequently. Reid Duke's data have a laughably small-n. Every article he presents basically tells us nothing about the metagame.
This has nothing to do with the comparison of UWb to UW presented.