Written by Glenn | 04 February 2011

Chalk up another holy shit moment to Jason Cole, who helped win the TourneyPlay 3-on-3 tournament at the bell last weekend. (UPDATE: OK I was told this match I'm about to break down is from a tournament in December. But the numbers on what happened are still amazing; the video barely got uploaded this week, which is why I was thinking this tournament just happened. I still stand with what I mentioned in the commentary). At this rate, he might own all of those insane precious moments in the Pacific Northwest in 2011. Last year his comeback ratio was one of the highest in the West Coast, so why not peg him to do it?

What made last weekend's this win exciting aside from him catching Mike Has Cookies just in time was the chances he had to do it. Based on similar situations, he was a virtual lock to lose, and yet, he didn't.

One of the stats I started tracking this year was combo rate. I figured it's the most fool-proof way of determining who had the best chance of winning. So far, it's the best determining factor. I went to my new database and looked at 642 rounds that were either semifinals or finals matches.

When someone has more combos than his or her opponent in a round, the win percentage has been .818. 

Hit the jump for a deeper breakdown...


Based on those rounds of play, when a player has more combos than his or her opponent, here's what the win percentage looks like ...

Combos	Wins	Win %
1	73	.682
2	77	.875
3	47	.940
4	21	.913
5	5	1.000
6	1	1.000

For the most part, more combos means a higher chance of winning (oh look, here comes Captain Obvious flying in, let me tug on his cape). This is in an overall situation. In specific situations (4-2 combos, 3-1 combos), the win percentages are tighter, and I'll get into that in a second.

In the final round of the final game of the tournament, Mike Has Cookies started off with two combos, but then Cole got two to even the round. Trailing late, Mike put in two more combos to even things up and get it down to its hair-raising final moments. Cole won despite having two less combos than Mike, two to four.

Combos in relation to wins are the only thing at the moment where I'm able to track win probability as they accumulate. Therefore, based on previous situations, here's how the win probability stacked up as the round was happening ...

Mike	Cole	Win prob.
1	0	.588 Mike
2	0	.912 Mike
2	1	.688 Mike
2	2	Even
3	2	.744 Mike
4	2	.900 Mike

In a 4-2 situation of recent championship and semifinal contests, players have won nine out of 10 times. The odds don't get any better. Mike had things in his favor. But then again, this is Jason Cole, the noted comeback extraodinaire. 

Here's where the mind gets blown. Not only did Cole do that in the clinching round, he did that also in the previous one, where he was about to be ousted. It didn't seem like much because he had that round more in hand (won with about 50 percent remaining).

Statistically the whole finale is a holy shit moment because Cole bucked the trend TWICE, getting out-comboed 7-4 in the final two rounds. And oh by the way, he landed one EX move in eight tries in those two rounds (.125 average) compared to Mike, who was 3 for 10 (.300).

In the two games Cole played in the finals -- he didn't play in the first finals -- he was out-comboed 14-8. Yet, he won both. That's a testament to him holding on through some adverse situations. Because at that rate, his opponents were getting one more combo per round. 

I said it a while back, that you can't turn away from a Jason Cole match no matter what the situation. Last Saturday, he provided more evidence.

Video of the two finals matches, provided by Team Khaos Gaming, is below.