In this article, I continue where I left off from my last article. We learned that it takes around 486 Takeaways per additional win. Teams as a whole rarely top 800 Takeaways in a season, and players who lead the league usually can get around 90 in a season. Even apart of the unreliability and subjectivity of how Takeaways are recorded, our research shows that it’s unwise for a team to target a player solely on his Takeaway skills since they won’t make much of an impact.
The less fun cousin of the Takeaway is the Giveaway. Takeaways for players are cool in that you are the one that forced the turnover. Giveaways are like unforced errors in tennis. You had the time to keep the play going and string together and additional pass, but you had a brain fart and just gave the other team the puck. Those Giveaways certainly aren’t helping your team, and today we’re going to investigate how much they cost you.
I did the same thing as Takeaways. I looked at play by play logs from the past two seasons of regular season game and tracked how soon after a Giveaway a shot for and shot against occurred. If there was a stoppage of play, I would “reset the clock” to zero and wait for another Giveaway. Below are the per game SF and SA values for two seasons worth of data:
For easy reference, here was the graph for Takeaways:
Fortunately, I think this passes the general sniff test. I’m only a little concerned about that spike at the 2-second mark, but can’t find anything in my code that would suggest any bugs. It’s tough to get a shot off one second after a team turns over the puck to you, so maybe two seconds is the sweet spot, especially when teams turnover the puck in the defensive zone.
There were 39,776 Giveaways over the last two full seasons (16.2 per game). Summing the shot differential value (SF-SA) for each second on the Giveaway graph above and dividing by 16.2 we get -0.19 shots per additional Takeaway. Multiply by 0.089 = -0.017 goals per Giveaway (or -1.7 Goals per 100 Giveaways). Three goals per point and two points per win means we take 6 / (-0.017) = -351.1 Giveaways per additional win.
Teams as a whole who lead the league in Giveaways usually come close to 1000. Players who lead the league often top 100 in a season. You’ll also find some great D-men usually leading the league (P.K. Subban, Erik Karlsson, etc.). Like Takeaways, Giveaways matter a little, but these stats when isolated shouldn’t be heavily considered when looking at a player’s contributions toward winning.
Recalling that it takes around 486 Takeaways for an additional win, you may ask why there’s such a disparity in value between the two. One reason is that most Giveaways occur from D-men. I’d have to do more analysis on where most of them take place on the ice, but if defensemen are giving up the puck, then naturally the opponent is in a better position to get a scoring chance since there are likely less defenders in their path. Another reason is that Giveaways contain more of a surprise element to them. If a player is getting pressured, there’s a higher chance of a Takeaway occurring. When a Giveaway occurs, there’s less notice that it’s coming, giving the team claiming the turnover more to work with before the team now on defense can set up.