Just the Stats

Just the stats was a statistical column on SportsIllustrated.com I co-wrote with Chris Apple during the 2001-02 season. Chris is a programmer and helped with the data compilation, whereas I came up with the methodologies (or applied existing ones) and did most of the writing. While I've mentioned the column before, I've never established an archive that is easily accessible. In this period of downtime, I plan to republish some of my other older works, such as my original pythagorean work - I examined the NHL back to Day 1, my attendance database, and a paper I wrote on AA pro realignment while I was working on my MBA (while not directly relevant, there are some points valid to junior hockey). Because the column built upon previous columns, I've listed them below in the chronological order in which they were written. I think it will help, since some things like my pythagorean applications require some foundational undersatnding to start. Since there's 17 essays, think of it as a semester of lectures. Hey, what did you expect from someone who's in academia? A new look at scoring effectiveness (9/30/01) Scoring has always been recorded in the simplest fashion possible. Someone gets the goal, a player or two gets an assist, and from there we move on. There's traditionally been only two ways to measure this: the straight tally (Pavel Bure scored 59 goals last season), and dividing that tally by the number of games played (playing 82 games, he averaged 0.72 goals/game). But we can do better... Would Have, Could Have, Should Have (10/15/01) Bill James, the Dean of Sabermetrics, has a simple tool to measure if a team over- or under-achieved for a given season based on the total runs scored and allowed. Ideally, a team that scores 800 runs and allows 800 runs should finish with a winning percentage at exactly .500. But what about teams who score 850 and allow 700? James came up with the following equation to handle this... A goal by any other name (10/22/01) All men may be created equal, but not all goals are. In a previous column, we examined the effects of scoring and ice time vs. points per games to measure efficiency of goal scoring. Yet another way to measure efficiency is to look at the situation in which goals are scored. I don't think anyone would argue that the most important goal is often the first one and in this the age of goaltenders like Dominik Hasek and Patrick Roy, one goal can mean the difference between victory and defeat. To better see the effect of scoring, we can measure "weighted goals," where we take the goals scored and revalue them based on the difference that goal makes in the score of the game. Advanced pythagorean methods (10/29/01) One of the problems with inventing a new science is that a lot of time is spent elevating the understanding of those being instructed. This week's column is not meant to intentionally come off sounding like a lecture, but in a way it is. A couple of weeks ago we introduced Bill James' Pythagorean equation and applied it to hockey. However, its uses are not limited to final regular season GF and GA for entire teams. Goals can be broken down into even strength and special teams. Then the equation can be applied again. The equation can also be applied to individual skaters, but that's another column. All assists are not the same (11/5/01) In a previous column we took a look at the value of weighting goals to measure the effectiveness of the scoring of a player. This same principle can be applied to the assists that result in these goals to look at the overall efficiency of the people creating the plays. Physical play... (11/13/01) If anyone can give us a valid reason that penalty minutes should be valued in fantasy leagues and pools, we'd like to hear it. A quick survey of the major online pools out there shows that every one gives some form of credit for minutes served, and frankly, we don't know why. If it's to give some measure of value to a physical player, it's the wrong way to do it. What's the value of saying the magic words and going to the box for 10 minutes? Up or Down with +/- (11/19/01) Is there anyone out there with an undying love for the +/- stat? No? We think everyone out there has some sort of issue with it. Many think it's unfair to penalize a player for the actions of his other teammates, but is not and argument since players benefit just as easily under the same circumstances. Besides, hockey is a team sport; so most individual stats are indirectly linked to the actions of one's teammates. Pythagorean equation tests a player's true impact (11/26/01) Last week, we explained how the +/- system doesn't go far enough to explain the impact of a player on the ice, in that two players each with +2 could have very different impacts towards a team's chance of winning. Instead, we are using some of the same raw data combined with ice time to compute a player's actual winning (or losing) impact on a team. Weighting goaltender performance (12/03/01) In previous columns, we have looked at weighting the goals and assists of scorers and playmakers in the NHL. We take the difference of the score when the goal was scored and value that goal based on a specific formula, previously discussed in 'A goal by any other name', then look at the difference between straight goal scoring and weighted performance. Now, what we will do is look at the reverse, weighting the performance measurements of goaltenders. We'll look at both the overall weighted goals against and the average weighted goals against for goaltender, and the results might surprise you. Weighting everyone the same (12/10/01) In this week's column we're expanding our vision to look at the backups. Just as a quick refresher and for those of you that didn't get a chance to read last week's column, we take the difference of the score when the goal was score and value that goal based on a specific formula, previously discussed in 'A goal by any other name', then we look at the difference between straight goal scoring and weighted performance. It's a matter of timing (12/17/01) They say that everything is a matter of timing ... it's no more true than in the game of hockey. The last minute save preserving a precious win, the last minute goal to give a needed two points in the playoff race. So who then are the early scorers versus the late scorers? When do most teams take penalties? Perhaps we can learn a little from the timing of events in the NHL. New Year's Resolutions (1/4/02) We will try to not miss any deadlines. We will try to come up with some meaningful numbers relating possession and territory to winning. We will try not to neglect goalie stats and methods as much as we have been thus far this season. Weighted saves, anyone? Playoffs projections from Pythagorean Performance (1/9/02) In a previous column, we've shown how the Pythagorean Method can be used to evaluate team performance based on their goals scored and allowed by comparing the derived theoretical winning percentage to the actual. To do this, we use the following equation... Scoring and special teams (1/18/02) This week we wanted to look at the effect that special team situations have on scoring in the NHL. Does killing penalties really hurt an offensive players scoring? Does a team truly benefit from being on the power play or hurt by being shorthanded in terms of overall scoring? There are many intangibles to special teams, such as player fatigue, losing a scorer to a penalty, but do special team situations really effect scoring? Situational pythagorean wins and losses (1/23/02) In an earlier column, we showed how goals can be broken down into even strength and special teams and the method applied to each piece of the data. From there we can weigh each piece based on overall impact, and with that we can assign actual W-L records for each team based on their even strength or special teams play. Fantasy Foolishness (1/30/02) Would someone please tell us who came up with the systems for fantasy hockey? The whole point is to pick players from around the NHL for a team, and then run a season based on the stats they accumulate in real life, right? Well, if that's the case, why are all these leagues using bogus statistical methods right and left?