clock menu more-arrow no yes mobile

Filed under:

Takeaways: Fun With FancyStats

This week's Takeaways are of a different flavor than most previous installments. Enough has been written elsewhere about the Stadium Series, so I'm going to take a look at some season-long trends instead of just those of the last week.

Kevork Djansezian

1. The Ducks still have the highest PDO (What's that?) in the league at 103.7, and their Corsi for Percentage (What's that?) is 15th at 50.2% (Extra Skater). I do believe that advanced stats are useful tools for analyzing the game and that they have predictive power, but I also believe the Ducks are legitimate Cup contenders this year. How can that be? A big part of it is the fact that I am a very biased Ducks fan who wants to believe they are contenders.

According this brilliant book, humans tend to form beliefs for subjective, emotional reasons and then manipulate the facts to support their beliefs. So it makes sense that I'm looking for loopholes in the numbers that make the Ducks look good. My latest harebrained idea has to do with score effect. Teams that are leading tend to have worse possession numbers, and teams that are trailing tend to have better ones. For a more thorough explanation of score effect, read this.

This phenomenon exists because the trailing team plays like they need a goal — because they do — and the leading team plays like they do not need a goal — because they don't. If a hypothetical team is playing with boatloads of confidence — okay, fine, I'll drop that ruse. If the 2013-14 Anaheim Ducks are playing with boatloads of confidence, they won't always feel desperate for a goal, even when they're tied or down by one. If they're not desperate, they won't play desperate, and their possession numbers will reflect that.

I'll admit that I've just laid out a very shaky theory. I haven't backed it up with any evidence besides the faulty train of thought that goes: The Ducks are good, advanced stats say they aren't, therefore advanced stats are missing something. Here's another idea.

2. It isn't completely impossible that, if possession were calculated by using a stopwatch to record how much of the game each team spent with the puck on their sticks in the offensive zone rather than by counting shot attempts, the Ducks' stats would look better than they do now.

Maybe Corsi and Fenwick underestimate teams that cycle the puck down low and look for higher percentage plays. Maybe teams like the Kings have good stats because they throw pucks and bodies at the net whenever they get the chance, and teams like the Ducks end up with high PDO numbers because they don't shoot from everywhere and they have plenty of scoring skill.

Take a look at their respective rosters. You'd expect the Ducks to have a higher shooting percentage than the Kings, wouldn't you? Maybe not. Maybe I'm crazy. Maybe I'm blinded by my emotional attachment to one team and I can't believe that their success is a phantom. But maybe there is a reason they're sitting seven points above the closest team with less than 30 games remaining on the schedule.

3. When the Boston Bruins won the Stanley Cup in 2011, they had the second highest regular season PDO close in the league at 101.8, and only the twelfth best CF% close at 51.1% (Hockey Analysis). Likewise, the Cup-winning Pittsburgh Penguins of 2008-09 had the second highest PDO close at 102.4 and the 16th best CF% at 49.2% (Hockey Analysis). They are the only two anomalies since 2007-08, when's data begins. Every other Cup team has had a middle-of-the-pack PDO and a top three CF% in the league. If the Ducks maintain their advanced stats ranking and win the Stanley Cup, they will be an exception to the rule, but not an exception that hasn't happened before.

Now for some more traditional Takeaways:

4. Mark Fistric is better than Bryan Allen. This probably became apparent a long time ago, but I wanted to wait until I had seen Fistric play for a long time before I made this judgement. Sample size and all that jazz. My ideal defensive lineup (barring a trade) looks like this:

Cam FowlerBen Lovejoy

Francois BeaucheminHampus Lindholm

Mark Fistric — Sami Vatanen

Scratched: Bryan Allen, Luca Sbisa

5. I'm very curious to see how the Ducks, particularly Jonas Hiller, play this week. The outdoor game was an emotional event, and it isn't hard to envision a let-down of sorts. Hiller admitted he had trouble focusing on the Winnipeg game last Tuesday, but he was the best player at Dodgers Stadium on Saturday.

6. I absolutely love the Ducks' outdoor game uniforms. If you're going to go orange, might as well go all orange. I hope we haven't seen the last of them.