Did you see the previous article specifically examining parity? I had a separate piece here on the blog that discussed the measure of parity that I used. There, I simply looked at a measure for parity for each season, and then divided the years into eras - pre-expansion, expansion, salary cap, etc. There was some arbitrariness, but the difference in the measures of parity was quite striking, I thought. At one point I did actually run a regression where the measure of parity was the dependent variable and I included dummies for the presence of the salary cap, expansion, etc. I didn't end up writing up those results (which had the salary cap having a statistically significant effect, although I forget the magnitude) because I didn't think it offered much that couldn't be seen looking at the numbers themselves.

As for your theory, I think you exactly right. The reason (or at least one of the reasons) we have a reverse order draft is to help bad teams get better. This is a different notion of parity. I previously looked at within-season parity, but the idea of bad teams improving from one year to the next is across-season parity. So while the new lottery will undoubtedly have some effect on tanking (and therefore within-season parity), it will also surely have some impact on across-season parity. It's hard to say exactly what that trade-off is, but it's certainly not negligible.

So, while I think that using the lottery for the first 3 spots might help (but maybe not - I guess we'll see), I would think that a lottery for all the spots would be a bad idea - for exactly the reasons you state.

Cheers,
Phil

I would love to understand better how you accounted for the salary cap's effect on parity.

I'd also be interested to hear your thoughts on next year's revised draft format (lottery for the first 3 spots). I applaud the NHL for continuing to find new ways of producing parity. However, I have a theory that the more randomized the draft order will have an inverse effect on parity.

My theory is based on the notion that the salary cap (though creating a more equal playing field for small market teams) has also significantly reduced the effectiveness of free agency as a means for a bad team to get better. If we assume that free agency has little to offer bad teams, than the draft becomes the only means of improvement. If bad teams become less efficient at improving through the draft (because the most valuable picks are more difficult to intentionally acquire) than perhaps rebuilds take even longer?

Rich Nash - 113
V Tarasenko - 110
A Ovechkin - 107
T Johnson - 106
M Pacioretty - 105
N Kucherov - 103
P Kane - 100
S Stamkos - 95
J Pavelski - 93
F Forsberg - 91 (rookie?)
J Tavares - 89
T Seguin - 88
SIDNEY CROSBY - 84 perhaps the 13th best player in the NHL?
10 more players
E Malkin - 74
*Phil Kessel, although 26th on the points leader chart has a Plus/Minus of -20, 17 points worse than any other top player, so Wayne you may be disillusioned with Toroto's star (and possibly why the Leafs struggle so much).

Just sayin'