Betting the NHL playoffs
A SIMPLE GUIDE ON HOW TO SUCCESSFULLY BET THE NHL PLAYOFFS
With the 2017 NHL playoffs set to begin tomorrow night, I figured I would revisit my dataset from the last 9 seasons (regular season and playoffs included) to see if I could find a winning strategy.
Some of you may have already read my previous article entitled “Betting NHL teams on a losing streak facing a team on a winning streak” (if not, I strongly encourage you to read it: https://www.professormj.com/pages/betting-nhl-teams-on-a-losing-streak). My initial hypothesis was that the public tends to overreact to recent results, thus over betting teams riding a winning streak facing an opponent coming off a series of losses. This study presented a method that yields a 15% ROI (Return On Investment).
Now, I wanted to verify whether I could find a profitable strategy for the playoffs based on the same assumption. For example, if a given team won the first game of the series but lost the next three, should we bet on them since most of the public will probably overreact to the outcome of the past three games?
For the sake of this article, the “higher seed” means the team that owns home-ice advantage. Games 1-2-5-7 are played at the higher seed’s arena, while games 3-4-6 are played at the lower seed’s home.
The data includes a total of 708 games, which corresponds to all playoff games between the 2007/2008 and 2015/2016 seasons. I have decided to take out the 2012/2013 playoff games because the lockout shortened the season, which may have had an impact on how things went and I did not want to introduce a potential bias in the data.
A word of caution before we jump into the main content: several findings below seem promising, but we often end up with fairly small sample sizes so you should take it into consideration before betting your house based on the results presented here. I’m not saying the results are worthless either, but please keep this fact in mind.
All right, so here is what I did: I bet on all higher/lower seeds in game #x of the series by breaking it up into the “current series status” (for example, betting game #3 by breaking it up into the three possible scenarios: the higher seed leading 2-0, the series tied 1-1 or the lower seed leading 2-0).
Let’s start right away by betting game #2 of the series:
Several studies have shown that it is best, generally speaking, to bet road underdogs as opposed to any of the other three possibilities (home underdogs, home favorites, road favorites). You lose less money, on average, by betting road underdogs. But I certainly did not expect to find a POSITIVE profit simply by betting the lower seed in game #2 of all 120 series in the dataset!! The ROI is twice better when the lower seed won the first game, which goes against my initial hypothesis that people overreact to recent games.
Let’s move on to game #3 of the series:
The only profitable strategy occurs when betting the higher seed in series which are tied at one game apiece. The rationale might be that the higher seed puts more effort and energy in such games because of the danger of falling down 2-1 with the next game being played on the road again, whereas they could be complacent when leading 2-0. But then, shouldn’t they be playing with desperation when trailing the series 2-0? Perhaps their confidence is gone and the lower seed plays with passion in front of their home fans after winning the first two games of the series on the road. Who knows? I can’t say I would be very confident about betting the higher seed every time the series is tied 1-1 because I trust a lot more strategies that follow my initial intuition.
Here are the results pertaining to game #4 of the series:
I absolutely love what I’m seeing here!! That’s exactly what I was talking about just above: I have much more confidence when results back my initial theory. At the very beginning I said I wanted to verify if we should bet against the grain by taking the team that has struggled the most recently. In this case, when the higher seed leads the series 3-0 we find a positive profit from betting the lower seed, while when the lower seed leads the series 3-0 we make money by betting the higher seed. I could not have drawn it any better than this.
How about betting game #5 of the series?
Oh my God… It gets even better. Once again, things go in the exact direction I was looking for. If we had bet on the lower seed every single time it trailed the series 3-1 over the past 9 seasons (except the 2012/2013 season), we would have made an unbelievable 53.7% ROI!! As mentioned earlier, the sample size (39) is fairly small so we are probably overperforming: the true ROI is likely to be smaller, but signs are strong that it is a viable strategy. Now, if you had bet all higher seeds when trailing the series 3-1, you would have made a 17.4% ROI, which is still nice! I’m impressed by the 16-7 record in such circumstances.
Let’s now have a look at the results for game #6 of the series:
The numbers in the table above indicate to bet the lower seed when trailing the series 3-2. That goes in accordance with the initial hypothesis, so I like it even though the ROI is not necessarily spectacular (still well above 0% though).
At this point, let’s recap all 8 winning strategies:
- Game #2: bet the lower seed when trailing the series 1-0
- Game #2: bet the lower seed when leading the series 1-0
- Game #3: bet the higher seed when the series is tied 1-1
- Game #4: bet the lower seed when trailing the series 3-0
- Game #4: bet the higher seed when trailing the series 3-0
- Game #5: bet the lower seed when trailing the series 3-1
- Game #5: bet the higher seed when trailing the series 3-1
- Game #6: bet the lower seed when trailing the series 3-2
What if we break it up into specific “paths”? First, what do I mean by a “path”? Consider the case where the higher seed leads the series 3-2. We just saw that it is a good strategy to bet the lower seed in game #6. Is it even more profitable if the lower seed won the first two games and then lost the next three, i.e. with the following path: LS, LS, HS, HS, HS? According to my intuition, it should be (the public might overreact to the past three contests).
Let’s revisit some of the cases above to see what happens. We cannot run this type of analysis on strategies 1-2-4-5 described above because there is only one possible path. For example, the higher seed leading the series 3-0 can only happen in the following way: HS, HS, HS. So we’ll look at strategies 3-6-7-8 in more details.
STRATEGY 3: We should be betting the higher seed when the series is tied 1-1.
Exactly what I was hoping for: we make much more money by betting the higher seed when it lost its previous game. As a matter of fact, we don’t seem to be making any money at all when it is coming off a win.
STRATEGY 6: We should be betting the lower seed when down 1-3.
Still based on the reasoning that people put too much emphasis on recent results, we would expect to be making more money when the lower seed lost the past three games. Despite the pretty small sample sizes, that’s what we observe here as the highest ROI occurs for this specific path (87.1% ROI). But I believe the main thing to remember from the table above is we are making money no matter what the path was.
STRATEGY 7: We should be betting the higher seed when down 1-3.
It’s funny how we witness a similar pattern: all four possible paths lead to positive returns (again despite some even smaller sample sizes).
STRATEGY 8: We should be betting the lower seed when down 2-3.
Again, I’m thrilled with the results since they are in accordance with the initial intuition. Remember we are betting the lower seed, so we expect to make more money when the most recent games went to the higher seed. When the higher seed won the last two games, betting the lower seed leads to a 7-3 record with a +4.23$ profit (ROI = 42.3%). When the higher seed won the last game, betting the lower seed leads to an 18-9 record with an 8.03$ profit (ROI = 29.7%). Finally, when the lower seed won the last game we expect to make LESS money, which is exactly what happens: 9-12 record with a -3.84$ profit (ROI = -18.3%). So the final word here is to bet the lower seed when trailing the series 3-2, but only if it coming off a loss.
Let’s summarize the way you should approach the NHL playoffs from a betting perspective:
Bet the lower seed no matter the outcome of the 1st game.
Bet the higher seed if it won the 1st game and lost the 2nd game.
Bet the lower seed if it is trailing the series 3-0.
Bet the higher seed if it is trailing the series 3-0.
Bet the lower seed if it is trailing the series 3-1 (bet more if it lost the last 3).
Bet the higher seed if it is trailing the series 3-1.
Bet the lower seed if it is trailing the series 3-2 and it lost game #5.
I hope you appreciated this analysis whose objective was to make you a winner during the Stanley Cup playoffs! If you liked the article and would like to support all the work I do related to sports betting, you can make a donation here. But most importantly, thanks everyone for taking the time to read this report, I hope I was able to bring you some valuable insight!!