There were 22 coaching changes between 2000-01 and 2003-04 in which the old coach had at least 20 games and the new coach had at least 20 games. The old coach put up a combined record of 141-219-52-28; the new coach a record of 160-197-54-29. If you express that on an 82 game basis, it’s 26-41-5-10 as opposed to 30-37-5-10. 75 points instead of 67.
The goal differential doesn’t support it though. The old coach saw his teams score 1057 goals and allow 1306. The new coach saw his teams score 1068 goals and allow 1250 goals. -249 versus -182. It’s probably worth mentioning that a large part of the change is attributable to the Blues switch of Mike Kitchen for Joel Quenneville - that resulted in a 30 goal swing for the Blues, leaving only 37 goal differential spread over the other 21 coaching changes.
On an 82 game basis, we’re talking about 197/243 versus 199/233. That’s more like four points in the standings - over the course of an entire season - than eight. And again, most of that seems to be tied to the Blues.
When you consider that a coach is probably more likely to get fired when things seem to be going poorly and that, in small samples, results don’t mesh up too well with performance (results swing more wildly than performance, IMO), the distinction between the results obtained by the new boss and the results obtained by the old boss probably gets even smaller. The old coaches went 62-71-28 in 1 goal games; the new went 72-57-29. Given that one goal games are largely random, you really have to ask whether there’s actually any difference in result created by the new coach, or whether there’s some truth to the old saying that you’re never as bad as you look when you’re losing or as good as you look when you’re winning.
If people want to advance the “Fire the coach” argument with respect to MacT, well, it’s a free world. I don’t, however, think that the evidence particularly supports that it would have much of an immediate impact or that it would really do much of anything. Coaching replacement level in the NHL is high and, broadly, one coach will make largely the same decisions as the next. The areas where they differ are going to be areas in which they’re fiddling around the edges. I don’t see that there’s enough there and, if I was running an NHL franchise, I’d be hesitant to eat the cost of the coach’s contract when I strongly suspected it would make little, if any, difference.
When you consider that a coach is probably more likely to get fired when things seem to be going poorly and that, in small samples, results don’t mesh up too well with performance (results swing more wildly than performance, IMO), the distinction between the results obtained by the new boss and the results obtained by the old boss probably gets even smaller.
Funny you mention this; I was just looking at the 12-game stretch that got Tom Renney fired (in which the Rangers went 2-7-3).
Over that stretch, the Rangers outshot their opponents at EV by 25 (in total EV shot attempts they were +92), but shot 4.4% and their SV% was .894.
You can just about guarantee that Tortorella will get better results (despite the same shit happening last night), because those percentages won’t hold.
I’m agnostic on whether the change was a good idea in the grand scheme of things, but I think Sather’s public reasoning for it (needed more “fire” in the dressing room) is garbage. Unless ‘effort’, ‘compete level’, etc. are reflected in ability to cash in rather than the ability to create shots and chances. Which I think is the exact opposite of how most people would intuit it, including me.
On the other hand, the first coach got to hold training camp and implement a system from day one, while the new coach is adapting on the fly to either a system not of his devising or trying to get a team to adapt to a new system midway through the year.
Unless you’re the New Jersey Devils, it seems to me that firing a coach midway through the season (while posting an average record) is more about giving the new guy an idea of what he has to work with for next season, or removing a problem from the team (ie. a few years back Trent Yawney decided to ice all four lines for fifteen minutes a game - unsurprisingly, he was canned).
And just moving past the specifics and on to a broader level - are you really arguing that there is virtually no difference from one NHL coach to another in terms of helping his team win? Because that just seems wrong, intuitively.
I don’t know how easy it would be to place a dollar value on the positive (?) press from making a coaching change.
For the Oilers, assuming they miss the playoffs, is it possible that they feel the positive reception from the average fan, if they change coaches, makes them more money than sticking with what they have?
Given that one goal games are largely random
Largely. Does the coach have any control over the one goal? Who knows? What I do know is that in the last calendar year the Oilers have played five overtime games where the oppositon has tied the score with their goalie pulled, and Zero overtime games where the Oil have tied it under similar circumstances. If one wants to put some of that on coach’s decisions like putting out, as Staples mentions in the previous post, Eric Cole as a last minute left-winger in a defensive situation against Calgary in which the play came straight up his side of the ice, or in putting Ethan Moreau, Fernando Pisani and Ladi Smid out in the last minute with our own goalie pulled, well, I think one is justified in at least considering those facts and wondering WTF?
Just for giggles, where did the teams that fired the coach finish in the standings the year prior, and where were they when they fired the coach?
I can’t come up with nearly as many good reasons not to fire MacT as I can to fire him but I’m not sold that firing him would change much. This is a team that came out last night night and looked completely uninterested in playing a game where they could have jumped from 9th to 6th place with 2 points. That says an awful lot about the players in the room. It is always easier to fire the coach than the players but there seems to be plenty of blame to go around for the place we find ourselves in the season.
Not knowing anything about the sitch, my only concern would be that MacT becomes untouchable in Edmonton, so that even when there are reasons to fire him, people don’t see them. On a chart of longest serving head coaches for a team with no cups in that period, where does MacT fall? It seems like he’s been there forever. At some point, isn’t longevity with no cups a good enough reason to fire? I’m only calling for his head if they replace him with someone as equally debonnaire.
I think its time for MacT to go myself. I don’t think that the team would suddenly go 10 and 0 or that Penner would become Frank Mahovlich.
Canning him now might shake the team out of its fog enough to bump them up and into the playoffs.
But they won’t dump him until the summer anyway so the point is moot.
But then they should do it. The kids aren’t getting better, the team is stale, they have no identity. Tonight apparently Pouliot is back in and its likely that Nilsson will be out again, which is absolutely ridiculous imo.
Moreau gets a free ride.
Penner gets benched again.
Its maddening. And its not working.
Its strange because they are where I figured they would be but I think they could be six to eight points better. And I do think that a lot of it falls on the coach.
I’m largely agnostic on this coaching issue. I really like the style of MacT the team—some of his decisions this year seem strange to me, but I can’t coach an NHL team so I’m in no position to question his judgement. On the other hand, I think change is good from time to time and that MacT has been here a long time; as well, I’m not a fan of nepotism, and MacT is part of the nepotistic family of the Oilers that I would like to see slowly moved out. If you forced me to take a position on changing the coach right now, I’d oppose for emotional reasons.
MC79’s implicit argument seems to me to be that we should discount the win-loss changes and instead look at massaged (admittedly validly) data on the goal differential which shows no significant change.
I point out that win-loss is what we should ultimately care about. Comparing the goal differential before and after is fraught with assumptions about similarity between the before and after sample groups, which might not be valid after a coaching change. Changes in how goalies are handled, which lines play, special teams play, etc. are all highly subject to coaching. As an aggregate, maybe the combination of changes made doesn’t affect goal differential, but they do affect the win-loss. E.g. new coach plays the young kids in blowouts more, writing-off the lost game and investing the time in development. When winning, he plays the guys with good goal-differentials all out since they’re the guys who win him games. Old coach wouldn’t throw the towel in a lost game because he, personally, needed the win to keep his job, so the loss wasn’t so lopsided. When winning, the old coach would also favour his long-time soldiers and perhaps underplay new players who have a positive goal-differential.
I don’t know if any of this happens, and I don’t have the time or commitment you guys do to studying this. I’m just suspicious of how MC79 discounts the win-loss using the goal-differential, rather than understanding the change in win-loss by using goal differential (for which is what I would use it).
I wonder if the win-loss is all because of guilt on the part of players that they got their old coach fired, so they put 100% effort in for a short period of time that boosts their win-loss on the short terms that MC79 has looked. Perhaps breaking down the before-after into 10 game segments would shed some light on this.
I broke it down into four ten games segments and here’s what I came up with:
(20-11 games before firing)73-107-32-8 536 GF 646 GA
(10-1 game before firing) 68-112-20-20 521 GF 660 GA
(1-10 games post firing) 83-100-26-11 542 GF 623 GA
(11-20 games post firing) 77-97-28-18 526 GF 627 GA
If you look at 11-20 games pre-firing to 11-20 games post-firing, there really isn’t a lot of difference.
Also worth nothing: teams tend to play more home games in the 20 games following a firing 11.45/20 to 10.8/20 before a firing.
If you look at the home games by quarter it goes 5.5, 4.3, 4.9 and 5.5. I would expect that the disparities in record are at least partly explained by the schedule.
You missed a golden opportunity to quote a Who song in the post title.
Anyway, I figure that the impact of a coach firing really depends on circumstances. Sometimes a coach is just a victim of luck and public pressure, but sometimes the coach has bought and paid for his own liquidation. If players don’t know WTF they’re supposed to be doing, as was allegedly the case in Montreal when Carbo was turfed, well, that’s kind of a problem. If a bench is being horribly mismanaged, that’s also something where I would expect a real change in performance over the long-term.
I also wonder if the Rangers’ shockingly low percentages (or any bad percentages) can be caused by a team-wide slump that’s corrected by changing the coach. I’m not saying that “lack of fire” or whatever the hell is causing them to miss, but if guys are squeezing their sticks, or tired of putting up with Coach X, they might be (unintentionally) firing it at the logo more, or just a little too slow to glove that shot, as you often see in slumps. I realize that pounding the sports psychology drum isn’t a popular move around here, but I figured I’d throw that out there.
Oh shit, I didn’t even notice that this was a three week old post.