Thursday, November 10, 2005
Good coaching, continued
Cosh makes a relevant comment to my last post about Coach Roy and Coach Gretzky:
I was all set to write a long comment about my suspicion that good road wins and/or a good road record are strong evidence of good coaching, but then I thought I'd do a bit of analysis to see if this bears out at all.
I started with a hypothesis that closely resembles the conventional wisdom: "Success on the road comes from playing a tight system, minimizing chances against, and not making mistakes; success at home comes from high energy, creating lots of chances, and talent riding the home crowd." My corresponding assumption is that the former relates to good coaching (smart system, on-ice player discipline, etc.) while the latter has very little to do with good coaching (we can argue here, but I believe the role of coach as "motivator" in professional sports is highly overstated).
So, how to measure this? If my hypothesis is true, then well-coached, under-talented teams will have a (relatively) stronger road record, while poorly-coached, highly talented teams will have a noticeably stronger home record.
Instead of comparing home and road Points Earned (which always slant Home; only 6 teams in '03/04 earned more on the road than at home), I'll compare their Ranks, then take (HomeRk - Road Rk). A large positive number thus indicates a team (and coach?) which is very strong on the road relative to at home, and a large negative number indicates the opposite. All stats from the full 2003/04 season.
The Top 7:
If my hypothesis is false, what are the alternative explanations for home/road differential? The first, most obvious candidate is luck; even if there was no human influence on the differential, it would exist, although my math isn't good enough to figure out how much.
I think you can toss out the "great home crowd" factor: Boston and Carolina had weak home crowds, but Colorado and Toronto have loud, supportive crowds. Similarly with the bottom teams, the crowd might spur Detroit and Nashville to greater success, but that doesn't account for why the Islanders and Anaheim were such (relatively) great home teams.
If it's not coaching, and it's not luck, then there is some other quality, measurable or not, that the Top 7 teams have in common (and the Bottom 7 don't). Ideas?
(Math footnote: tie ranks were averaged, e.g. 3 teams tied for 7th/8th/9th would all receive a rank of 8th, for the purposes of the arithmetic operation noted above. Ranks shown next to the team are not averaged, e.g. those same 3 teams would all be shown as T7th.)
Gretzky will have total freedom to implement a plan, but it won't do him any good unless he actually has one.
I was all set to write a long comment about my suspicion that good road wins and/or a good road record are strong evidence of good coaching, but then I thought I'd do a bit of analysis to see if this bears out at all.
I started with a hypothesis that closely resembles the conventional wisdom: "Success on the road comes from playing a tight system, minimizing chances against, and not making mistakes; success at home comes from high energy, creating lots of chances, and talent riding the home crowd." My corresponding assumption is that the former relates to good coaching (smart system, on-ice player discipline, etc.) while the latter has very little to do with good coaching (we can argue here, but I believe the role of coach as "motivator" in professional sports is highly overstated).
So, how to measure this? If my hypothesis is true, then well-coached, under-talented teams will have a (relatively) stronger road record, while poorly-coached, highly talented teams will have a noticeably stronger home record.
Instead of comparing home and road Points Earned (which always slant Home; only 6 teams in '03/04 earned more on the road than at home), I'll compare their Ranks, then take (HomeRk - Road Rk). A large positive number thus indicates a team (and coach?) which is very strong on the road relative to at home, and a large negative number indicates the opposite. All stats from the full 2003/04 season.
The Top 7:
- Boston - Home T18th, Road 1st, Diff. +17.5, Coach Mike Sullivan
- Colorado - Home 20th, Road T2nd, Diff. +17.5, Coach Bob Hartley/Joel Quenneville
- Carolina - Home 26th, Road 14th, Diff. +12.0, Coach Peter Laviolette
- Toronto - Home T13th, Road T2nd, Diff. +12.0, Coach Pat Quinn
- Vancouver - Home T13th, Road 4th, Diff. +10.5, Coach Marc Crawford
- Los Angeles - Home T23rd, Road 13th, Diff. +10.5, Coach Andy Murray
- Calgary - Home 17th, Road T9th, Diff. +7.5, Coach Darryl Sutter
- Dallas - Home 2nd, Road 18th, Diff. -16.0, Coach Dave Tippett
- Islanders - Home T5th, Road T19th, Diff. -15.5, Coach Steve Stirling
- Anaheim - Home T14th, Road 26th, Diff. -11.5, Coach Mike Babcock
- Detroit - Home 1st, Road 11th, Diff. -10.0, Coach Dave Lewis
- St. Louis - Home T8th, Road T15th, Diff. -7.5, Coach Joel Quenneville/Mike Kitchen
- Nashville - Home T8th, Road T15th, Diff. -7.5, Coach Barry Trotz
- Buffalo - Home T13th, Road T19th, Diff. -6.5, Coach Lindy Ruff
If my hypothesis is false, what are the alternative explanations for home/road differential? The first, most obvious candidate is luck; even if there was no human influence on the differential, it would exist, although my math isn't good enough to figure out how much.
I think you can toss out the "great home crowd" factor: Boston and Carolina had weak home crowds, but Colorado and Toronto have loud, supportive crowds. Similarly with the bottom teams, the crowd might spur Detroit and Nashville to greater success, but that doesn't account for why the Islanders and Anaheim were such (relatively) great home teams.
If it's not coaching, and it's not luck, then there is some other quality, measurable or not, that the Top 7 teams have in common (and the Bottom 7 don't). Ideas?
(Math footnote: tie ranks were averaged, e.g. 3 teams tied for 7th/8th/9th would all receive a rank of 8th, for the purposes of the arithmetic operation noted above. Ranks shown next to the team are not averaged, e.g. those same 3 teams would all be shown as T7th.)
# posted by Matt @ 10:32 am 
Comments:
Just one point. Sometimes homecrowds can be a disadvantage. I am thinking particularly of the Montreal crowd where it seems the pressure can be unique--particularly on young players.
I was going to say that length of travel and duration of roadtrips might have some effect (i.e., teams in the West get screwed because of the brutal travel schedule), but your numbers don't seem to back it up.
Actually, they might -- 10 of the bottom 14 teams (with a worse road ranking than home) are from the Western Conference; you could guess that this is due to the longer roadtrips (bad) and longer homestands (good). But the Top 10 has 5 teams from each conference, then the next 6 are all from the East.
Also (FYI Tim), Montreal is T10th Home and 12th Road.
I would argue that a sign of good coaching this season is decreasing your goals against (per game) this year vs. the 2003/04 season when scoring is up leaguewide. It requires bringing in a solid defensive scheme.
4 teams have done this so far this year. Detroit, Ottawa, NY Rangers and Phoenix.
Is it reasonable to suggest goals against are dropping in Phoenix while rising almost everywhere else without a good defensive scheme? Wouldn't that have to come from coaching?
I'm starting to be won over. I think Gretzky will be a good coach. I was very skeptical when the season began - considering he had never coached at any level and now had an NHL job.
Post a Comment
<< Home
Just one point. Sometimes homecrowds can be a disadvantage. I am thinking particularly of the Montreal crowd where it seems the pressure can be unique--particularly on young players.
I was going to say that length of travel and duration of roadtrips might have some effect (i.e., teams in the West get screwed because of the brutal travel schedule), but your numbers don't seem to back it up.
Actually, they might -- 10 of the bottom 14 teams (with a worse road ranking than home) are from the Western Conference; you could guess that this is due to the longer roadtrips (bad) and longer homestands (good). But the Top 10 has 5 teams from each conference, then the next 6 are all from the East.
Also (FYI Tim), Montreal is T10th Home and 12th Road.
I would argue that a sign of good coaching this season is decreasing your goals against (per game) this year vs. the 2003/04 season when scoring is up leaguewide. It requires bringing in a solid defensive scheme.
4 teams have done this so far this year. Detroit, Ottawa, NY Rangers and Phoenix.
Is it reasonable to suggest goals against are dropping in Phoenix while rising almost everywhere else without a good defensive scheme? Wouldn't that have to come from coaching?
I'm starting to be won over. I think Gretzky will be a good coach. I was very skeptical when the season began - considering he had never coached at any level and now had an NHL job.
Post a Comment
<< Home
