TEAM SCORING CHANCES VS CORSI
“ScoringChance%” is the percentage of the total even strength scoring chances in the game that were owned by the Oilers. So if, over twenty games, the Oilers racked up 200 EV scoring chances, and the opposition tallied 300 EV scoring chances … then the Oilers would lay claim to 40% of the scoring chances over the stretch (200 of 500).
“Corsi%” is calculated using the same reasoning, as is EV “Goals%”.
Corsi% is our best guess at the extent of territorial advantage held
Quite clearly Corsi% (or any of the shots metrics for that matter) ends up giving us a clear indication of scoring chance percentage. Using 20 game rolling averages as above, they are almost on top of one another.
On the year as a whole:
· ScoringChance% = 47.7
· Corsi% = 47.8%
· Goals% = 51.0%
While corsi and scoring chances tend to converge in the long run, the two do not necessarily correspond over smaller sample sizes, such as a single game, or even a collection of games.
To give a concrete example, if two teams are relatively evenly matched, and one team establishes a lead early in the game, and retains that lead for the remainder of the game, there’s a good chance that the trailing team will establish a sizable corsi advantage by the end of the game. However, the scoring chances will likely be more or less evenly distributed (after all, the teams are evenly matched).
I suppose my point is that, relative to corsi, scoring chances are more reflective of the balance of play over shorter periods, and therein lies its utility (at least from my standpoint)
CLOSE RANGE SHOTS PREDICTIONS
Shot differential is a common metric for assessing teams. Do we do better if we focus on the close-range shots?
A lot of today’s hockey analysis relies on the importance of shot differential in predicting future success.
Team shooting and save percentages often run hot for a stretch, but in the long run that success is rarely sustainable and so shot differential is usually a better predictor than looking at goals or wins.
In some cases, various measures that account for shot quality might do a slightly better job of predicting future results than simple shot differential.
He found that at a certain point in the season, a team’s fraction of the shots from close range (within 25 feet) was a better predictor of its future outcomes than its overall shot differential was (Corsi or Fenwick). He also found stronger correlations to future results from his Expected Goals metrics that weights each shot based on its location and type.
I worked out the correlation between even strength shot differential (or ES close-range shot differential) and a team’s future points – not just for a single day, but following each game of the NHL season.
So instead of just a table with correlations on a couple of days, I can produce a whole curve showing how the correlation changed over the course of the season
After the season is about 1/5 of the way done, we get superior predictions of future point totals by focusing on just the close-range shots than if we include all shots on goal. It’s not just a single date; it was true for the majority of the year.
Close-range shots did a bit better last year than in the older data, but I wouldn’t have looked at this plot and concluded it was a better measure than simple shot differential.
I find myself unwilling to make predictions based on the close-range shot measures at this point in time
Measurer ^2
Corsi 0.62
Fenwick 0.65
Expected Goals 0.73
Goals 0.79
Shots inside 25 feet 0.73
NIELSON NUMBERS VS CORSI
what are Neilson Numbers. They are a version of +/- that attributes a “plus” to a player who is seen to directly enable a scoring chance for, and a “minus” to a player who makes a mistake causing a scoring chance against
In theory, Neilson numbers are great. They are obviously better than conventional +/- both because they don’t award marks to players who have nothing to do with a play and because they are based on scoring chances rather than goals, which is good because shooting success is largely – but obviously not totally – luck-based.
So what’s wrong with them? Well first of all, anything to do with scoring chances is incredibly subjective. The most objective way to track them is to count any shots from the home plate area that spans from the goal posts to the faceoff dots to the top of the circles, but that really just tracks “close in shots” rather than scoring chances.
The second major issue with these numbers is that they don’t take into account the probabilistic nature of hockey, not just when it comes to shooting, but also driving play up the ice. Players that are able to make defensive stops, to transition to offense quickly, to generally lead his team to good results without necessarily making the flashy play, are going to be severely undervalued using these metrics.
Corsi (or on-ice shot attempt differentials, if anybody is new to fancystats) certainly has its flaws. Often players will get a “plus” or “minus” for a shot attempt they had nothing to do with, but the advantage is that because the sample is so large, this variance after long enough will get filtered out.
While something like corsi has its flaws, the main advantages to it are that it is mostly objective in terms of tracking, and that it focuses on results over a large sample. In essence, it says ‘I don’t really care how you make your team score more goals when you’re on the ice than the other team, I just care THAT you do.’ Obviously, this is once you take contextual factors into account.
That last point is really important. Process is obviously important, but in the end if one can prove with statistical significance that player makes his team better, then the question of why is of second-hand importance.
BETTER THAN CORSI – SCORING CHANCES
(Jan 2015)
Better Than Corsi: Scoring Chances More Accurately Predict Future Goals
Since the 2005-06 NHL season, the percentage of Scoring Chances For (SCF%) is a better predictor of future Goals For (GF%) than Corsi For (CF%) is for individual players.
We find that:
· For every one-percentage-point increase in SCF%, future GF% is expected to rise by about 0.41 percentage points, holding all other variables constant.
· For every one-percentage-point increase in CF%, future GF% is expected to rise by about 0.22 percentage points, holding all other variables constant.
PREDICTABILITY DIFFERENCE FORWARDS VS DEFENSE
(Jan 2015)
This is an update to our previous post on the best metrics for predicting player performance. Here, we split the analysis out by position (forwards vs. defensemen).
· For forwards, SCF% is the best predictor of future GF% of the metrics we tested
· For forwards, CF% is a better predictor of future GF% than is FF%, but for defensemen, the opposite is true.
· For defensemen, FF% is the best predictor of future GF% (with SCF% finishing a close second) of the metrics we tested.
· SCF% is a much better predictor of future GF% for forwards than it is for defensemen.
For Forwards:
· Both SCF% and CF% have statistically significant associations with future GF%.
· In general, future GF% is more accurately predicted for forwards than it is for defensemen.
· For every one-percentage-point increase in SCF%, future GF% is expected to rise by about 0.49 percentage points, holding all other variables constant.
· For every one-percentage-point increase in CF%, future GF% is expected to rise by about 0.21 percentage points, holding all other variables constant.
Interestingly, the magnitude of the SCF% coefficient increased for forwards, indicating that SCF% is a better predictor of future GF% for forwards than it is for defensemen in this analysis.
Note that we also repeated this analysis using FF% instead of CF%, but FF% was found to have an insignificant effect on future GF% when accounting for SCF% in the model (results not shown). This is very interesting: CF% was found to be significant even after accounting for SCF%, while FF% was not. This may indicate that the additional information included in CF% — blocked shots, for and against — is driving some of the metric’s predictability of future GF%. Our hypothesis is that blocked shots against (i.e. shot attempts taken by the opposition at the player’s goal) are driving the effect here, since forwards do a lot of shot-blocking at the points in the defensive zone.
Season-to-season, the metric with the highest past-vs-future correlation for forwards varies. Recall that across all seasons, SCF% has the highest past-vs-future correlation. This seems to be backed up by the graph, where SCF% is highest by a relatively large margin in 4 of 9 seasons.
For Defensemen (SCF vs Corsi)
· Both SCF% and CF% have statistically significant associations with future GF%.
· For every one-percentage-point increase in SCF%, future GF% is expected to rise by about 0.31 percentage points, holding all other variables constant.
· For every one-percentage-point increase in CF%, future GF% is expected to rise by about 0.22 percentage points, holding all other variables constant.
In other words, SCF% has a more substantial effect* on future GF% than does CF% in this analysis.
For Defensemen (SCF vs Fenwick):
· Both SCF% and FF% have statistically significant associations with future GF%.
· For every one-percentage-point increase in SCF%, future GF% is expected to rise by about 0.25 percentage points, holding all other variables constant.
· For every one-percentage-point increase in CF%, future GF% is expected to rise by about 0.30 percentage points, holding all other variables constant.
Interestingly, SCF% is less predictive of future GF% for defensemen than it is for forwards, and FF% is the superior predictor of future GF% for defensemen by a small margin in this analysis.
Season-to-season, the metric with the highest past-vs-future correlation for defensemen varies quite a bit. Recall, though, that across all seasons, FF% has the highest past-vs-future correlation. This seems to be backed up by the graph, where FF% appears to be a bit more consistent from season to season than other metrics.
EXAMINING SCORING CHANCE DATA
(Jan 2015)
The average probability of a goal given the type and locations, and the consideration of team defense, we have these conditions for a “scoring chance”:
· In the low danger zone, unblocked rebounds and rush shots only
· In the medium danger zone, all unblocked shots.
· In the high danger zone, all shot attempts (since blocked shots taken here may be more representative of more “wide-open nets”, though we don’t know this for sure.)
In essence the measure is attempting to identify a higher percentage attempt, that has an increased likelihood of resulting in a goal. Thus we can surmise that attempts meeting these conditions are of greater value, and this has been assessed as being the case specifically for year-to-year relationships for individual skaters and description of past results.
All-Situation Scoring Chance For Percentage (SCF%) correlates more highly to past Goals For percentage (GF%) at the team level (r2 = 0.3805) than Unadjusted Fenwick For percentage (FF%) does (r2 = 0.3658), but at a lower rate than any of the Score Adjusted metrics available. SCF% also has lower repeatability at the team level than CF% and FF%, likely due to the relative rarity of the events being recorded.
SCF% predicts future year GF% to a higher degree than Score Adjusted CF% and FF% for forwards. For defenders, Score Adjusted FF% predicts future year GF% to a higher degree than Score Adjusted CF% or SCF%. This is likely due to the direct impact on scoring attempts Forwards seem to have, while defenders tend to be put in a more passive position as far as offense is concerned.
One of the main objectives with all of this in terms of future assessment is of course an analysis of in-season predictivity. We wish to know which of these statistics is best in-season at informing us about future outcomes. We can recall in my previous posting relating to the predictivity of various shot attempt metrics that we had reached a point where the current state of affairs indicated that Score Adjustment of Corsi and Fenwick improved the in-season predictive power over Score Close and Raw measures substantially.
In a fashion similar to that undertaken by Micah-Blake McCurdy I explored the in-season predictivity of 5v5 SCF% vs GF% and All Situation SCF% vs GF%.
Interestingly the results are less powerful than we might hope based on the other information gleaned to date.
It appears that the In-Season Predictivity of SCF% for GF% at both 5v5 and in All-Situations maxes out around the 20-25 game mark. All-Situation SCF% reaches 19.2% while at 5v5 it peaks at about 18.4%. Neither value is particularly impressive in comparison to the predictive power we currently have in the form of Score Adjusted 5v5 CF% or FF%.
As you can see in the following graph, McCurdy found that 5v5 Score Adjusted CF% reached a maximum in predictivity around the 20-25 game mark of approximately 30%. The substantial increase is likely due to the fact that Corsi and Fenwick accrue at a much faster rate than Scoring Chances and information about the teams we are assessing is that much greater at an earlier stage. In essence we are back to the sample size problem.
The value of scoring chances in terms of description of goal outcomes is obvious in comparison to raw shot attempt counts. That being said we aren’t accumulating enough data quickly enough to improve predictivity of future outcomes. We also know that Unadjusted SCF% doesn’t out-perform Score Adjusted CF% or FF% at the moment in terms of describing past results either.
Given these results, it looks like next steps amount to exploring a weighted Shot Attempt model that includes some Score and Venue Adjustments. Weighting attempts based upon their location of origin on the ice, and the time dynamics of the event (i.e. rebounds or rush attempts) should improve the detail of the information contained in each event. By Score Adjusting and still including all events, we should accrue data at a high enough rate to theoretically improve upon the results we are seeing from Score Adjusted CF% or FF%.
ARE SCORING CHANCES BETTER THAN CORSI
Scoring chances are more predictive than Corsi.
There were two things that made Corsi so valuable: it was highly repeatable (ie. If you know a team’s Corsi in one part of the season, you can predict its Corsi in their remaining games very well) and it was better than goals at predicting future goals (ie. If you know a team’s Corsi and its goal differential, you can get a more accurate prediction of the team’s future goal differential by using Corsi).
Two things are clear here: Corsi is better than high-danger chances at predicting future goal scoring, and scoring chances are better than either of them.
I think it’s fair to say both that the scoring chance model from Natural Stat Trick is an improvement on Corsi and that it’s pretty comparable to existing expected goal models
I think the evidence for preferring scoring chances to Corsi is pretty strong. While SCF% does show a bit less autocorrelation, it’s still a pretty stable statistic, and over the past 4 seasons it has shown a better ability than Corsi to predict future goal scoring.
There is no reason to use high-danger chances. It’s not a good stat.
Comentarios