Mon Mar 21 03:36pm EDT
(Ed. Note: Typically found on Sunday, Stat Nerd is where we occasionally obsess over hockey numbers like a Dungeon Master obsessing over the level of his warrior elf. Here's Matt Barr, formerly of LCS: Guide To Hockey and Trolleytracks and now blogging hockey at Kertwang.me.)
Craig Ramsay, whose Atlanta Thrashers' postseason hopes are on life support after an 8-2 shellacking from the Buffalo Sabres over the weekend, has started backup goaltender Chris Mason(notes) three times since March 1 and brought him in for relief of starter Ondrej Pavelec(notes) twice, with mixed results.
Mason held down the fort the first week of March for an injured Pavelec, and won one game and lost two. He got credit for an amazing come-from-behind win against the Flyers March 12 after starting the third period down 3-0. Against Buffalo, he couldn't stanch the bleeding, starting the second down 3-1, then allowing five more goals.
"Mixed results" is a kind description for what teams on the playoff bubble get when they don't have one starting goalie they ride till he drops down the stretch. We looked at 62 "bubble" teams in the five full seasons since the lockout. We classified a team as being on the bubble if it was in eighth place in its conference, or within five points either way, on March 1 of the season.
For 2005-06, whose season ended a week later than usual, we made it March 8. (Average games played by 2005-06 teams from March 8 through the end of the season worked out to 19.78; for the other years from March 1 on, 18.72. Close enough.) Thirty-five of the 62 made the playoffs.
We looked at total minutes played by whichever goalie for the teams played the most minutes. Here's some stuff we learned.
Average starting goalie minutes played for bubble teams that made the playoffs: 916:13.
Average for non-playoff teams: 833:01.
Average minutes played per team game for playoff teams: 48:31.
Average minutes played per team game for non-playoff teams: 44:11.
Ranking the 62 teams in order of starting goalie minutes played, most to least, you find that of the top half -- that is, the 31 highest goalie minutes played totals -- 64.5 percent (20 of 31) made the playoffs.
Of the bottom 31 (15 of 31), 48.4 percent made the playoffs. The percentage doesn't change if you rank by minutes played per team game instead.
Naturally, your starting goalie has to perform. Ranked by March/April Save Percentage, two-thirds of the top half made the playoffs; 42 percent of the bottom half did. Of course, if your starting goalie isn't up to snuff, you won't play him as much, so chicken or egg?
Either way, the takeaway is: If your bubble team has a starting goalie playing well, and he plays a lot, you're in better shape when it comes to making the playoffs than if you're playing the hot hand, or have injury problems.
We looked at other factors, such as special teams. Rank the 62 teams by March/April power play conversion percentage, and you see 61.3 percent of the top half making the playoffs; 51.6 percent of the bottom half. Penalty killing?; 67.7 percent of the top half, 45.2 percent of the bottom.
A rough "special teams" rank — PP percentage plus PK percentage — shows that 70.1 percent of the teams with the "best" special teams down the stretch make the playoffs, and 41.9 percent of the "worst" do.
We looked at shot differential, too: 34 of the 62 teams had more March/April shots on goal than they allowed, and 28 had a negative differential -- 70.6 percent of teams with a positive differential made the playoffs; 39.3 percent of those with a negative differential did.
Conventional wisdom wouldn't argue with outstanding starting goaltending, sound special teams, and shot discipline as ingredients for a playoff team. It's instructive to see them quantified for teams fighting for their playoff lives down the stretch.
Now, how do bubble teams do once they're in the playoffs?
Another bit of conventional wisdom goes that end-of-season momentum is important for a playoff team. That turns out to be not quite the case.
To look at how teams that had to scratch and claw their way into the playoffs did once they got there, we defined "bubble teams" a little more granularly. We're not just interested in whether making the playoffs was likely or not, we also want to know how hard a team had to fight, and for how long. Here's what we did:
If a team was in eighth or within five points either way six weeks before the end of the regular season, we gave it three "bubble points." Four weeks out, if a team was in eighth or within four points, it got four "bubble points." Then, any team in eighth or within three points two weeks out got five "bubble points."
We made one adjustment. If a team was both (a) within five with six weeks to go, and (b) within three with two weeks to go, we gave it the four "middle" points automatically. We doubt any team felt safe about its playoff chances a month out if it was on the bubble with six weeks to go and with two weeks to go.
So. Does having to fight for your life down the stretch give you an advantage in the playoffs?
First, comparing the bubble teams themselves. Do you do better the harder and longer you have to fight to get in? The data don't really say one way or the other.
Teams with 12 or more "bubble points" won an average of 4.58 playoff games once they made it. Teams with seven "bubble points," who didn't have as hard a time making the postseason, won an average of 5.89 games. But just when you think it's better not to have to fight tooth and nail every night, you find that teams with three to five "bubble points," the ones who had the comparatively easiest times, won only 2.13 playoff games. Overall, teams with seven or fewer "bubble points" won 4.12, to 4.58 for 12 or more.
When you take all the bubble teams together and compare them to playoff teams that were never on the bubble — never in any real danger of not making the playoffs — you see that it's better to be good than to have bubble-forged "momentum."
Forty-one bubble teams overall the last five years, ignoring "bubble points," won 180 playoff games, or an average of 4.39 per year. Thirty nine teams who were never on the bubble won 245 games, or 6.28 per year.
One "bubble team," the Pittsburgh Penguins in 2008-09, won the Stanley Cup. Only three of 10 Stanley Cup Final teams have come from the "bubble": the Edmonton Oilers in 2005-06 and Philadelphia Flyers last year, in addition to the Cup champion Penguins.
The moral of the story is: Stay out of the playoff bubble to begin with. But once there, ride your goalie hard and put him away wet.