Is Past OT Performance a Good Predictor of Future OT Performance?
by Hawerchuk on Nov 25, 2009 9:00 AM EST
in Analysis
Email
PrintCommenter Saskhab had some concerns about the assumption that OT performance is primarily luck. In particular: "I don’t think we can use 5 on 5 play as the predictor for 4 on 4 play."
I think, in the long-run, this is true. If a team played 10,000 minutes of sudden-death OT 5-on-5 and 10,000 minutes of sudden-death OT 4-on-4, I think we'd find that the 4-on-4 play was a much more reliable indicator of future OT performance. But as is often the case in sports, we don't have large observed samples like that - what we have is a very small sample of the event in question (4-on-4 OT) and a larger sample of a dissimilar event (5-on-5 play in regulation.) To give you an example, the Montreal Canadiens have played a total of 60 minutes at 4-on-4 so far this season, compared to 1076 minutes at 5-on-5 - 60 minutes is not a lot of time to prove your abilities.
I looked at each team's shot differential in 2008-09 and divided shots into two groups: 1) 5-on-5 shots taken in tied games; 2) 4-on-4 shots taken in OT (when the score is also tied, obviously). I then compared each team's performance in odd and even games, and their performance in the first half of the season and the second half. Here are the results:
| 5-on-5, O/E | 5-on-5, 1/2 | 4-on-4, O/E | 4-on-4, 1/2 | |
| R^2 | 14.9% | 13.6% | 2.4% | 0% |
| Regression to the Mean | 61.5% | 57.6% | 84.5% | 100% |
Odd/Even games are abbreviated (O/E) and First/Second games are abbreviated (1/2). As you can see, 5-on-5 shot differential is a better predictor of future 4-on-4 performance - in fact, 4-on-4 shot differential in the first half of the season has no relationship with 2nd-half shot differential. This is primarily because teams played 25 times as much at 5-on-5 as they did at 4-on-4, which gives us a much better assessment of their true talent than the limited amount of time they spent at 4-on-4.
13 comments
|
0 recs |
Do you like this story?
Comments
OK, I can accept that. I guess 4 on 4 is such a small consideration that it’s like projecting how a team does in the PP in the first round of the playoffs and projecting their success for the remainder of the playoffs based on that small sample. I still don’t think 5 on 5 has much to do with 4 on 4, and that was my main problem in that argument, but you’re right… it’s more likely to produce results we can live with than OT stats at ANY point of the year (because the sample just doesn’t ever get large enough to be satisfactory).
And, like someone pointed out… the Habs have scored 4 times on 23 scoring chances in OT, while not allowing a single one on 15 chances against. That SHOULD produce a winning result, but they also are bound to lose some as a 1.000% SV% is unsustainable (therefore going 4-0-5 in 4 on 4 situations is a real anomoly).
Hockey blogging can't get any flatter.
I think you can think of it this way, for illustration: the Habs have played 60 minutes of overtime in total, scoring 4 goals and allowing zero. That’s not unlike winning a single hockey game by a score of 4-0. Sure, games like that happen, but what are the odds of another such shutout over the next game, even if your team is very good?
The Habs are a weird team this year. They’re battered, they lose games they should win, they win games they should lose, they allow over 50 shots to Nashville, they come up with three-goal comebacks with a salary floor roster. Really hard to tell what they really are.
Methodology
Methodology question…
You took each team’s total shots (i.e. – corsi) at 5-on-5 w/game tied in the first half and made that your x (estimator). Then you made y (response) each team’s own performance in corsi at 5-on-5 w/game tied in the second half?
I ask because
1) Wouldn’t that only show correlation between 5-on-5 to 5-on-5. (As well as 4-on-4 to 4-on-4). But you said:
[i]“As you can see, 5-on-5 shot differential is a better predictor of future 4-on-4 performance”[/i]
so maybe I have your methodology wrong and you did in fact correlate 5-on-5 to 4-on-4.
2) If my initial guess at your methodology is correct, I’m surprised you got such low correlations. I just did the following:
I took each team’s corsi at ES wgame tied (the data is from TOI though, so i think Vic includes 5-on-5 and 4-on-4 together. Not sure if he includes OT.) through the first 52 games of 08-09 (weird number of games, I know, but it happens to be the data I had) and compared it to each team’s corsi at ES w/game tied for the final 30 games of 08-09.
I get r^2 of .475 (r = .69)
Incidentally, I then did the same thing for goal differential at ES w/game tied and get:
r^2 = .03 (r = .17)
ps – i also think Vic strips out EN goals
by sunnymehta.com on Nov 25, 2009 2:27 PM EST up reply actions
You know the game is broken when ...
Canucks try to make their point
Team has only been able to reach two overtime games this season
- – today’s Edmonton Journal
… the Canucks — billed as a tight-checking, hard-working club — have only been able to get two games to overtime this season. That’s the fewest in the NHL and a sore point for a team that, on average, has got more than on-quarter of their games to overtime during the past four years … “It’s really surprising that we still have a zero in that [OTL] column,” Alex Burrows said. We’re more than a quarter of the way into the season and we only have been able to get to OT twice. Those points are huge."
Note the emphasis not so much on OTL points but simply on getting to overtime. Do you think they are on to the system? Frustrated cuz the dice haven’t rolled their way to this point, but it’s not like they’re not trying to play tie games. That’s how you keep up with the Joneses in Gary Bettman’s NHL.
Writer for The Copper & Blue and primary shareholder of Zorg Industries
btw, I totally feel the frustration you’ve been venting in your last couple of posts here, Bruce.
Even regardless of its effect on actual results, the 2-1-1-0 point system is intrinsically detrimental to the quality of hockey. The incentive structure reinforces awful traits – collusion, playing to tie, minimizing all scoring chances – basically everything that is against the spirit of competition!
I suspect the main reason the league likes it is because they feel it creates parity, and parity generates money. What we need to do is put together a full-scale study on the exact value of the 2-1-1-0 point system compared to a 3-2-1-0 point system, in terms of expected revenue generation for the NHL/its teams. If we can prove that the current system is not only a drag on quality of hockey, but is also not really creating much more revenue than a 3-2-1-0 system would, then we’d be onto something.
Even regardless of its effect on actual results, the 2-1-1-0 point system is intrinsically detrimental to the quality of hockey. The incentive structure reinforces awful traits – collusion, playing to tie, minimizing all scoring chances – basically everything that is against the spirit of competition!
Doesn’t that just say it all. And yes, Sunny, my frustration is pretty extreme. As I’ve put it in big bold letters in a couple of my posts, the Free Lunch point system does nothing less than compromise the integrity of the sport. If that doesn’t set off the warning bells to these dumbf*cks, you wonder what will. Do they really think 18,000 people want to fork over top dollar to watch another tie game where both teams play as if they’re protecting the lead?
Writer for The Copper & Blue and primary shareholder of Zorg Industries
by Bruce McCurdy on Nov 25, 2009 5:57 PM EST up reply actions
I think we’re always going to have a struggle here between the good teams and the bad teams (who opposed OT in 1983.) Initially OT incentivized winning, but teams gradually trended back to preferring a 90% shot at a tie (and 1 point on average) than a 60% shot at a win. The same thing happened after OT went to 4-on-4.
It’s ironic that playing more minutes – which should give the better team more opportunities to win – turns out to be better for the weaker one.
OT favors the weaker team — Is it because OT is sudden death?
If they always played the full five minutes, perhaps it would rebalance things a bit back towards the better team as they would be more likely to come back from an early OT deficit (although not much, given the relative scarcity of goals).
Just getting to OT favours the weaker team
Anything else is a bonus. Maybe they only average 1.4 points per OT game instead of the mean 1.5, but that still kicks the shit out of the < 1.0 they will average in regulation games.
Writer for The Copper & Blue and primary shareholder of Zorg Industries
by Bruce McCurdy on Nov 25, 2009 7:20 PM EST up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
