It’s Endgame Moves Week for Overwatch, wrapping up the cross-franchise educational content series that previously featured Opening Moves Week and Midgame Moves Week. This time we’re focusing on the all-important final stages of each map or match that often mean the difference between victory and defeat. Yesterday we explained the origins and implications of the C9, and today we’re breaking down some overtime numbers.

End Game Moves: C9, Explained

On the dreaded C9—where it comes from, why it happens, and how to avoid it.

What is “endgame” in Overwatch? Is it the final teamfight in a map, or in a round? For brevity, I will define the term “endgame” as a very specific slice of data: any teamfight that either begins during overtime or contains the beginning of overtime. These are the teamfights that unequivocally determine whether a round is won or lost, as attackers fight to continue their attack while defenders battle to end the round. Every overtime teamfight, by definition, has the potential to be the “endgame.”

In these high-pressure moments, a single final blow or death can determine the winner of an entire match—and sometimes it’s five final blows in a row, like Jae-Hyeok “Carpe” Lee’s famous “winnable” moment. With these highlight reels in mind, I created two charts, diving into the final blows and deaths that occur during overtime in teamfights.

Let’s start with final blows:

Overtime final blows per overtime teamfight plotted against the delta between overtime teamfight win rate and global teamfight win rate (min. 25 overtime final blows). Data includes matches from Stages 3 and 4, including Stage Finals, and the postseason. Highlighted players were picked to show the general boundaries of the data.

Above, you can see a chart of overtime final blows (OTFB) per overtime teamfight vs. overtime teamfight win rate* in matches from Stage 3 onwards. I plotted these two stats against each other in order to evaluate players more holistically, where a single stat would fail. For example, Carpe and Jeong-Woo “Sayaplayer” Ha—both primary Widowmaker players—attained similar OTFBs per overtime teamfight (0.84 for Carpe, 0.70 for Sayaplayer). However, Carpe performed much better in overtime situations than Sayaplayer, as evidenced by his overtime teamfight win rate delta.

In other words, despite scoring only 0.14 more final blows per overtime teamfight than Sayaplayer, the difference between Carpe’s overtime teamfight win rate (59%) and his global win rate (47%) was +12%. Sayaplayer’s was, unfortunately, -6%.

* Teamfight wins are teamfights where a team scored more final blows than deaths. I’ve kept this simple so that more complicated future analyses can be built off it.

This brings up a chicken vs. egg conundrum. Do these numbers suggest that when Carpe secured OTFBs, they directly led to teamfight wins for the Philadelphia Fusion? Was Sayaplayer as good as Carpe last season, but dragged down by his Florida Mayhem teammates in overtime situations? Or was Sayaplayer the ultimate “garbage time” DPS, only securing final blows when they didn’t matter?

It’s worth noting that Carpe does not have the highest raw overtime win rate (that’s Do-Hyeon “Pine” Kim, with 70%), but I think we can agree these endgame stats tell a story that we already know to be true: Carpe turns it on when the game is on the line. His +12% win rate in overtime (relative to all teamfights) is the best in the league.

Onto our second chart:

OT Deaths - Player.png
Overtime deaths per overtime teamfight plotted against the delta between overtime teamfight win rate and global teamfight Win Rate (min. 25 overtime deaths). Data includes matches from Stages 3 and 4, including Stage Finals, and the postseason. Highlighted players were picked to show the general boundaries of the data.

Above I have plotted overtime deaths (OTD) per overtime teamfight against the same overtime teamfight win rate delta in matches from Stage 3 onward. The players who stand out the most to me here were members of the Dallas Fuel. If you had told me at the end of Stage 2 that not only would Dallas make the Stage 4 playoffs, but they would in some ways appear more clutch than the likes of Jong-Ryeol “Saebyeolbe” Park, Jun-Young “Profit” Park, or Terence “Soon” Tarlier, I’d have laughed. But the Fuel didn’t make it with offense; they were winning with solid defense, thanks to the arrival of main tank Min-Seok “OGE” Son and Pongphop “Mickie” Rattanasangchod’s early mastery of Brigitte, which gave them the ability to clutch out wins in Overwatch’s overtime endgames.

Ben “CaptainPlanet” Trautman is the statistics producer for the Overwatch League Global Broadcast. Follow him on Twitter!