Round 4: DING!!!

Wrote this a few days but didn't post it. I realized that what I was proposing was far too massive of an undertaking to be completed here so I saved, closed, and drank beer instead.

A much better third round, as will certainly be recognized in the comments section by fawning minions. I hate saying it, but I would've taken the Pens in 5 if I'd known Timonen wasn't going to be playing (even though he later did), simply because of how well he played against Washington and Ovechkin. Dallas won the expected 4th game at home but where did they come off winning Game 5 in Detroit? Rude.

Finals? Pittsburgh in six. I have no confidence in this. None. I don't even have any rationale either, because in my mind both goalies are the same, Pittsburgh's scoring is a tad higher-end than Detroit but the Wings defense and puck control is better. Powerplays both have lots of pop, but since Detroit is still missing Franzen (likely to miss the first game or two), I'm leaning towards the Pens. With little confidence.

I was challenged by Unimpressed Reader Matt to follow up on my off-the-cuff comments about most over-rated athletes. The charge was that it's easy to come up with over-rated athletes, but under-rated athletes would make for a more interesting list. Over and under-rated are pretty subjective, if they play on your team they're under-rated, if they play for the New York Yankees they're over-rated. A true list of under-rated players should, obviously, juxtapose a player's performance against their perceived value. Player performance is pretty easy to quantify, certainly not perfectly, but there's no end to the number of statistics you can pull and manipulate. Yet how would one quantify a player's perceived value? The first things that come to my mind are the market they play in, since more media coverage will yield a more exaggerated characterization of value, both good and bad. The second is obviously salary. That's a commentary on our society isn't it? That we assume that salary is an appropriate measure of a person's value? It's only after that I come around to the number of awards a player has won, peer-voted or otherwise, which really ought to be the first thought but instead comes in a delayed third.

The final product would be the result of the following equation. Take a list of players from a league of choice, categorize them by position, and rank them according to their statistical performance. For example, points per game by defensemen in the 2007-2008 NHL season, or the team's record in games they played in vs. games they did not play. That's the Performance ranking. The Hype rankings would be a composite of awards (maybe), salary, and market size. Rank the same players by salary, awards won, and market size. Salary is a straightforward numeric ranking, awards won would include not just end of year awards but also awards from the Hockey News, Globe and Mail, TSN, stuff like that. Market size, I propose, would be determined by calculating the average NHL city size, then for each player on the list ranking them against the mean. I'm not going to do that here, but let's say the average NHL city is one million people. Anyone playing for the New York Rangers, the largest NHL market at 8.2 million people, would receive 802% more attention than the "average" city-size. So, players would be ranked on a plus, or minus, ranking in order to sloppily account for player hype and attention received in proportion to their market size. You'd do the same with salary, take the average and rank players according to their relation to the mean. Awards are where I'm stuck, not sure whether to include them in Performance or Hype. Some are strictly for performance, like the Art Ross for most points, but others, like best defenseman or goalie, are much more subjective. I don't know about that one yet.

You'd take each of the three (or two) Hype Quantifiers (Yes? No? I like it, catchy and faux academic), reverse rank them so that the top ranked players receive the most points (so if there's 100 defenseman, the highest paid receives 100 points). This is just so that the players at the top of the list have the highest Hype Quantifier numbers, making them stand out more. Add the two (or three) sets of points together for each player, then re-rank the players according to Hype Quantifier results and compare them with your Performance rankings. The players with the biggest discrepancy are your most over-rated or most over-rated, depending on which list they're highest on. That's my rudimentary formula for now. It's pretty flexible since it would allow a person to use any stat they want to measure performance since points obviously is not always the best performance indicator.

Quantifying perception is pretty hard. When talking about hype, most of it comes from the media. In my last post I linked to that article writing about Dion Phaneuf is over-rated. Well, where does that rating come from? Pierre McGuire mostly, but the rest of the media has fallen in love with him too. That's something that should go into the Hype Quantifier somehow, but I don't see how. If you included some sort of points system that counts the number of times a player is mentioned in the media, Sean Avery would probably be near the top. A media ranking wouldn't really work since they simply cover a story rather than performance. Often that's both, but suspensions and poor performance are stories too.

As mentioned at the top, I'm not going to do this list now. That's simply HOW I'd like to do it and since I'm finding there's too much subjectivity in my own blog analysis, I'm trying to find ways to formulate some actual answers and results. No stat is perfect or tells the whole story, but this is a first step. I might not even do the list, just keep tweaking the formula according to Unimpressed Reader comments. I'll call it the Brian Urlacher Performance and Hype Composite.