In what is becoming a tradition, I will attempt to analyze some data while drinking. Matt's argument was that the single goalie spot exposed managers to the vagaries of game scheduling and starter-backup cycling. A skillful manager, nurturing their team to greatness with special care for the G-spot, would suffer unfairly when their teams' schedule had few games for their stud G, or worse.
Presented below is a summary of (almost) all the match-ups in the 2010-2011 season. For each weekly match-up, I calculated the difference between the number of games played by the goalie of Team A and the number of games played by the goalie of Team B. The column represent the frequency of wins (light bars) or losses (dark bars) for Team A in the weekly match-ups.
Matt's proposal to add a second goalie spot would likely reduce the week-to-week variance in number of goalie-games. How much? I don't know. I could look at the games-played pattern for some or all of the skating positions, which would give an idea on the effect of scheduling (but not the backup issue). I can't seem to find a good source for those data from the league website (Kev?), so that might have to wait.
The current scoring system for goalies is balanced and fair, but I think the mechanics of the h-t-h system is distorting the ultimate value of goalies. A couple of weeks of bad luck with scheduling and a few backup starts and your team is in a big hole, even though your goalie is a 2.0 ppg. Let us design a system that rewards skillful drafting, crafty trading, and epic bullshit, rather than punishing teams for the vagaries of scheduling.
I support Matt's proposal to add a second goalie spot. I think it will diminish this games-started effect I describe here, and returning the focus of the match-up to the players and their performance. We will likely need to re-scale the goalie scoring to reflect the larger inputs of total positional scoring to team performance, but I think with the fantasy wizards McCartney at the helm, we will not be steered wrong.
This is great analysis, Ben, and one I'd never seen. It shows me that goalies are worth More than I thought they were, while I thought the complaint was that they were worth too little. My mistake.
ReplyDeleteMy concern is two fold, I think. With only 23 (by eye) of the 139 total events taking place outside the +/-1 game (16.5%), is this a real problem? I can see the dramatic increase in losses by giving up a goalie game, but the fact that it's not mirrored on the other side of zero (and that zero has a substantial difference) makes me think that this might be less causal than it appears (at the +/-1 game level, anyway). And if so, a full 83.5% of our weekly match ups are somewhat in balance, even if that one game difference is much greater than it would be for another position.
Second, I too would like to see the games variations for skaters (and I'll send you an email about the best place I can imagine for that data), because I'm not sure that increasing the number of goalies will decrease the summed effect of game variation. What I mean is that I'm not sure that the difference it will make in the size of the effect of each game played will not be compensated for (and perhaps made worse) by the difference made in the deviation of total number of games played per week by goalies.
Only 25 goalies played 50+ games last year (including awful goalies like Brian Elliot and Steve Mason), and we're proposing 28 starters (with the potential for teams to carry 3 or even 4 goalies at a time). So we're likely to see that chart expand to include weeks with 4, 5, or even 6 game differences between goalie pairs. Ideally, that means the chart is then shaped more like a U (more game difference = more causal difference in wins). But even if we achieve that statistically, it means that having good goalies becomes more important practically as a matter of heightened scarcity. Without good goalies, you're more likely to be on the wrong end of that U-curve every week. Currently, I suspect there are few if any teams that are consistently up or down in number of goalie games per week. In a two-goalie system, having more goalie games will not only be predictable, it will be necessary to win. In essence, I fear we trade statistical balance for practical balance.
Thanks Kev. I think the effect is very real even at the +1/-1 goalie-game differential. Those represented ~1/3 of all the matchups last year, and having a goalie-game-in-hand nearly doubled your odds of winning. You were almost twice as likely to win if you had that extra game. That effect is larger than I would have expected, and fascinating in its own right.
ReplyDeleteThe data from the skater position games-played would be very useful for exploring if adding a second goalie slot will be appropriate, as you argue. I really don't know if the second spot is the fix, and the skater data would be a good test of the idea. Given a trained-team of stats monkeys, we could actually simulate a range of scenarios using historical data. Now that I think about it, it is Friday, I do have a number of bottles of wine...
And you are right about the transformation this would have on the market for goalies. With expansion, we quickly start scrapping the bottom. Will that vigorous competition for high-quality G add to the league? Or will it just be annoying like running an autodraft team?
Dale's (I think it was Dale) suggestion to draft teams rather than players for G is interesting, and might avoid some of the some the variation issues. If we want to deal with the goalies-starts issue, we could also standardize goalie points to games per week, or something.
Hi Ben,
ReplyDeleteThat is a very interesting way to frame this debate, and very different than I had thought of it to this point. You make a good case that number of goalie games is a deciding factor, but I wonder if that is a reason to increase the number of goalies on a roster. We face similar problems with players, and I feel like it is part of good roster management throughout the season to make sure that you get the best goalie in a position to succeed.
We do the same thing with other positions - I have many defencemen (seven to be exact) who I will switch in and out of the lineup as their opportunities wax and wane. If one of my d gets number one pp minutes on his team, he will be in my lineup more regularly than a guy who gets number two minutes. I feel like it is the same situation with goalies. I only have one, so I am going to (for the foreseeable future) have to ride it out. But Andrew, who has drafted more effectively than me, has two, and he can switch one in or out as the games played dictates (by the way, does my dependence on qualitative data show my educational background vis-a-vis y'alls clearly, or what). I think the significance of your information is that we are challenged to make good strategic decisions when it comes to goalies every week.
As a side note, what is really interesting from both these posts is that they have shown, very clearly, that what we all assumed to be true at the draft was actually false. At the draft, we all proposed that goalies are not valuable enough, and that there is no positional scarcity for them. But in reality, there is tremendous scarcity, and they are clearly very valuable. It is an interesting insight, and it will change my strategy pretty dramatically going forward.
Dale
"At the draft, we all proposed that goalies are not valuable enough, and that there is no positional scarcity for them. But in reality, there is tremendous scarcity, and they are clearly very valuable."
ReplyDeleteTo continue from my other post in the first goalie debate blog post this was not what Dale told me through chat. To repeat so that you don't waste your time looking for my other pointless rantings...I said they are too valuable and spreading thin would be an issue, he told me that is what other people said.
Let me be clear, this is not an attack Dale thread. I am just worried that with people like Dale having to talk to 3-4 people at once, certain people not being present, and others maybe not paying attention (I am that guy many times) people may not be voting with a properly informed viewpoint. I realize that is why we pushed this out of the AGM for further discussion, but I believe my concerns are valid.
My experience from 2010-2011 season.
ReplyDeleteI keep two goalies hoping that both would be studs, one I could keep and the other trade to help upgrade my team. Unfortunately both Miller and Howard were middlings. This was for three reasons 1) Their teams played worse than in 2009-2010 2) They played worse. 3) Especially in the case of Miller, injuries, albeit, 1-2 games at a time, cost me too many points in many weeks.
My point...I likely don't have one. I am just trying to show that even with two goalies you can still get shafted. If I had to actually play both for points instead of trying to guess which one to play each week my situation would have been even worse. Now compound my issues by trying to do that with a second goalie that is likely a backup. Problem made even worse.
I guess I do have a point...I vote no.
Regarding the team versus goalie idea. I vote no to this as well. This protects people to much. Take Andrew's case. Why should he get rewarded if Luongo stinks by automatically getting Schneider? This situation should be an opportunity for someone to swoop in and finally get a goalie on a good team. If Andrew is worried about that he can always take CS as a bench player. I do not know Miller's backup, but he started to suck even if Buffalo got better this year, than same situation. You feel me?
ReplyDeleteOmar
Just reviewed my posts...sorry about the spelling and grammatical errors. I hope your eyes are OK.
ReplyDelete