Big Time Points+

Inspired by Mr. Plank's latest Sharktistical Analysis, Big Time Goals, I set out to try to improve it.

I completely agree that not all goals are equal. Potting one in the last five minutes of Tuesday's game against the Leafs wouldn't have helped the Sharks win any more. But sending one into the empty net in the final seconds of Friday night's game, during that ridiculous 6-on-4, would've put the nail in the coffin. Since neither of these would have tied the game or put the Sharks ahead, they both end us as a big 0 BTP.

Or consider this: Your team's down by 4, and you singlehandedly score 3 goals. But you don't get a chance to score the tying goal, because the opposing goalie mishandles the puck on a clear, and then does it again, giving your team the lead, and the BTGs to some yahoo defenseman, even though you were the reason he got that chance.

So I think it needs to be tweaked a little bit. My solution: FRACTIONS!

A game-tying point or a tie-breaking point remains the same value: 1. If your team ends regulation with a tie, your team gets one point, so you get one point for helping that happen. If your team ends the game with the lead, your team gets another point, so you get one point for helping that happen.

But if you're just running up the score, or you're down so far that the opposing team has suited up their clubhouse staff, then you don't get a full point. Similarly, you get more for putting your team up 4-2 than making the score 4-0. Right now, this is how I'm calculating it:

-If your team is losing, for every goal or assist you get: 1/(OpposingScore - YourTeamsNewScore + 1). So if they have 4 and we have 0, whoever scores a goal or gets an assist will get 1/(4-1+1), or 1/4 Big Time Point+. If it's 4-3 and you tie it up, you get 1/(4-4+1), or 1 BTP+.

-If it's tied or you're winning, you get 1/(YourTeamsNewScore - OpposingScore). Break a scoreless tie? 1/(1-0) = 1 whole BTP+. Pretending you're playing football? 1/(14-7) = a whole 1/7 of a BTP+.

Got it? A tying or tie-breaking goal is still the most valuable, but it doesn't overlook the steps it took to get there. It's certainly not a perfect formula (in particular I'd like to add a coefficient or exponent to the denominator to value Lead+1 and Tie-1 goals more), but I do think it's worthwhile. And now onto the stats...

Sharks 08-09 BTP+ totals, minimum 10 G+A (sorry Cheech):

1. Marleau 21.917
2. Thornton 21.167
3. Setoguchi 20.333
4. Boyle 18.033
5. Pavelski 16.083
6. Clowe 15.283
7. Vlasic 13.583
8. Blake 11.617
9. Ehrhoff 11.283
10. Michalek 10.117
11. Grier 6.033

No big surprises there. How about BTP+/P? What percentage of regular points turn into BTP+ points?

1. Vlasic 79.9%
2. Pavelski 76.59%
3. Marleau 75.57%
4. Boyle 75.14 %
5. Clowe 69.47%
6. Setoguchi 67.78%
7. Ehrhoff 66.37%
8. Thornton 66.15%
9. Michalek 63.23%
10. Grier 60.33%
    
11. Blake 55.32%

Vlasic jumps to the front here because of his huge BTA+ score. 12.83 BTA+ out of just 14 assists easily offsets his dismal 0.75 BTG+ for 3 goals. Even using the original BTP, he has a 12/17 ratio, meaning 70.59% of his goals and assists give the Sharks either the lead or a tie. Boyle is in a distant second with 62.5%, and Marleau leads the forwards with 58.6%.

With the original BTP/P, Boyle and Pavelski switch places, as do Setoguchi and Clowe, and Ehrhoff and Thornton.

The next step, I think, would be to factor in time. I'm thinking something along the lines of length until a particular goal is negated, i.e., a goal in the first minute of play that is the only one scored in the game is worth far more than a goal in the first minute of play that is answered five minutes later. Or perhaps ranking individual players not on percentage of Big Time Points to regular points, but BTP(+) to time on ice? I dunno... What do you guys think?