I've made some updates to the post based on recent results...
The phrase "must-win game" has been overused to the point that it's now almost a meaningless cliche. Below the jump, I'll describe what a "should-win game" is, why it's more useful analytically, and what it means for the Washington Capitals' playoff chances.
Through overuse, the term has come to be synonymous with "big game," "important game," or "game against a key rival." For a team that is fighting for a spot in the Stanley Cup Playoffs, I propose that "should-win game" is actually a more important term as it comes to understanding a team's playoff chances.
A should-win game is a game that a team ought to win based on the opponent or circumstances such as their position in the standings.
Should-win games are the contests a team expects to win in order to make the playoffs. They're the winnable games against weaker opponents. They're the home games or games against a team on the second night of back-to-back games. A should-win game isn't necessarily against a big rival. More likely as not, they're against teams toward the bottom of the standings. If a team loses a should-win game, they've got to make up those points against tougher opponents, which is far less probable.
Clearly, not all games are should-win games,"so where does that leave the remaining games? I'd argue that there are two other types of games: "Get-a-Point Games" and "F*ck-It Games."
Get-a-Point Games are the toss-ups. They're a game that a team could win, but the outcome is far from certain. In Get-a-Point games, escaping with a point by getting to overtime is a positive accomplishment. (By contrast, only getting a point out of a "should-win" game would be a disappointment.)
F*ck-It Games are exactly what they sound like - they're the games that a team does not expect to win. Games against tough opponents, games against good teams on the road, and games on second night of back-to-back games can all be F*ck-It Games.
For the sake of simplicity (and easier euphemisms) I'll abbreviate Should-Win Games as SG, Get-a-Point Games as GAP, and F*ck-It Games as FIT. With that in mind, let's take a look at the Caps' upcoming schedule and classify each game.
(Author's note - I've added Neil's percentages from the comments. Neil, if you've got more recent data than March 15, feel free to share and I'll update this again.)
- 3/13 @ New York Islanders - SW - WIN - This turned out well.
- 3/16 @ Winnipeg Jets - GAP - LOSS - (The Caps looked like they might steal a point in Winnipeg, but came up just short.)
- 3/18 @ Chicago Blackhawks - GAP - LOSS - (This was admittedly a long-shot game to get a point in, but the Caps could really use all the points they can get.)
- 3/19 @ Detroit Red Wings - FIT - WIN - (Well, that's why they play the games. I certainly didn't believe the Caps would have a snowball's chance in hell in this game.)
- 3/22 @ Philadephia Flyers - FIT (30.4%)
- 3/23 vs. Winnipeg Jets - SW (74.8% - This is the second of back-to-back games, but the Caps could effectively eliminate Winnepeg with a win. I debated making both this and the Philly GAPs.)
- 3/25 vs. Minnesota Wild - SW (75.3%)
- 3/27 vs. Buffalo Sabres - SW (76.2%)
- 3/29 @ Boston Bruins - FIT (24.9%)
- 3/31 vs. Montreal Canadiens - SW (60.9% - This looks suspiciously low compared to the probabilities for the Wild, Sabres and Panthers games. I'd guess the probability is closer to 75+%.)
- 4/2 @ Tampa Bay Lightning - GAP (35.0%)
- 4/5 vs. Florida Panthers - SW (68.2%)
- 4/7 @ New York Rangers - FIT (22.7% - This could turn into a GAP or SW if the Rangers have clinched the East and the Caps are still fighting for position.)
Thus far, the Caps have won an SW, lost two GAPs, and won one FIT. As Gould Old Days stated in the comments below:
... we can be fairly certain that they’ll lose at least one that they should win, and that they’ll win at least one that they should lose. It’s the nature of the endeavor.
Gouldie was totally correct, but I don't think that invalidates the analysis or the categorization. The Caps lost two points that they might have had a decent shot at acquiring in Winnipeg and Chicago (and they also lost the chance to all-but-eliminate the Jets), but they managed to pull off a Motor City Minor Miracle. Either way, my categories had them coming out of the first four games of this road trip with four points, and lo and behold - they came out with four points. (Albeit not quite how I would have expected.)
Using this eminently scientific breakdown, the Caps have five SWs, one GAP, and three FITs remaining as of 3/21/2012. Breaking that down further into wins, losses, and OT/SO losses, that means the Caps ought to get five wins, three losses and one OT/SO loss. That would give the Caps eleven points over the final nine games, and would be good for 91 points. Though close, 91 points would likely be enough to allow the Caps to squeak into the playoffs. Furthermore, by beating the Jets, Sabres and Panthers in head-to-head matchups, the Caps could help themselves by hurting their nearest competitors.
Although they did it in improbable fashion, the Caps came out of the first four games of the road trip with four points. If they can win the "should-win" games and get a point out of the "get-a-point" games (either by OT/SO losses, or by splitting those games 50/50 win/loss), they could set themselves up for a playoff spot. In addition, if the Caps can steal a point or two in one of the "f*ck-it" games (Thursday night in Philly would be a good place to start), they would likely have a playoff spot locked up. Unfortunately, the Panthers have run off a little hot streak, so the division may be out of reach.