Fancy Stats In Fantrax

[yougo]

I often hear that said about Hits, but many, many leagues still use it. And I think it is also true of Missed Blocks - very, very subjective. I deal with these kinds of measurements at work all the time. If a measurement is consistent but not necessarily accurate, it still gives you a basis for evaluation. In a week's time, it won't be very good, but over a season, the inconsistencies should average out. So I would say stats like these are better for Roto than H2H, where short term results can cost you wins.

Yes but go take a minute and look at the leader in giveaways in the NHL and you'll see all the nominees for last year Norris trophy.

Drew Doughty, 2015-16 Norris winner, ended with a -87 giveaway/takeaway ratio. Doesn't makes much sense if you ask me.

[finminer]

Doughty is definitely an anomaly - 0.10 ratio, -87 total with only 10 takeaways in 82 games. But that doesn't mean that the stat is that subjective or inaccurate. Interestingly, it doesn't look like there are many of the top 50 d-men who are positive (ratio 1.00 or more). But thinking that through, coming up the ice, they handle the puck more and quick forecheckers can take it away. On defense, the d-men are in the back and have fewer opportunities to get in wide open passing lanes; things are much more congested down low. Look at the forwards' ratios - many of the top 50 are above 1.00. I'd say that while there are certainly outliers, in general the numbers make sense. These are statistics - anything will happen that can. It kind of reminds me of point guards in basketball - they can be the best in the league, but since they handle the ball so much and throw so many passes, they are also the ones who have the highest turnover rates.