Oilers vs Panthers 2024 - True Odds Calculation

Ok, so, arguably, the title of the blog is a bit of click-bait. The 'true odds' of any situation is too complicated for mere mortals to determine. However, this is a head-to-head comparison of 20 (or 23) statistical categories, that should correlate with success in hockey.

The columns show the scores for each team, in the categories, over the 2023-24 regular season. The weighing factors come from a long time-trend of Stanley Cup playoffs for each category. They show the win percentage for each category, over this period, for the team with a better score in that category. Note that these win percentages are on a game-by-game basis for all playoff levels.

So, for example, home teams won 54% of games over the visitor. Similarly, the higher ranked team in the regular season won 55% of playoff games over the lower ranked team. The team with the higher goals per game in the regular season won 54% of their playoff games. And so on, for all the rest of the categories. The final three categories are ratios, based on some of the variables listed above them. One could argue that doing this just repeats information already in the comparisons. Or one could argue that the ratios actually do give additional insight into the situation. So, I gave both alternatives. A fully worked-out multivariate analysis would shed light on the matter, but that is for another time and blog.

Seq	Category Name	Edm Oilers	FLA Panthers	Weight (Win%)	Pct Edge	Edge Winner
1	Home Games vs Visitor	3	4	0.54	0.04	FLA
2	Rank (points)	104	110	0.55	0.05	FLA
3	Goals For per Game	3.56	3.23	0.54	0.04	EDM
4	Goals Against per Game	2.88	2.41	0.54	0.04	FLA
5	Power Play (PP) Goals For	64	63	0.53	0.03	EDM
6	PP Opportunities For	243	268	0.51	0.01	FLA
7	PP Success Pct	0.2634	0.2351	0.51	0.01	EDM
8	PP Against	53	51	0.53	0.03	FLA
9	PP Opp Against	258	291	0.49	-0.01	EDM
10	Penalty Kill	0.7946	0.8247	0.52	0.02	FLA
11	SH Goals For	7	8	0.51	0.01	FLA
12	SH Goals Against	5	9	0.49	-0.01	EDM
13	Pen Min Against Team/Game	9.5	13.6	0.51	0.01	EDM
14	Pen Min Against Opp/Game	8.5	13.5	0.51	0.01	FLA
15	Shots on Goal (Team)	2768	2764	0.53	0.03	EDM
16	Shot Pct (Goals Scored)	0.105	0.096	0.51	0.01	EDM
17	Shots on Goal (Opponent)	2569	2496	0.48	-0.02	EDM
18	Total Saves	2307	2279	0.48	-0.02	FLA
19	Save Pct (Goals Prevented)	0.898	0.913	0.53	0.03	FLA
20	Avg Age	29.4	29.5	0.53	0.03	EDM

21	Shots Ratio: Team/Opp	1.08	1.11	0.53	0.03	FLA
22	Goals Ratio: For Against	1.24	1.34	0.54	0.04	FLA
23	Ratio: PP/(1-PK)	1.28	1.34	0.52	0.02	FLA

Next are some relatively simple prediction models, based on the above data.

Model 1 - Simple Category Winner Models

The method below totals the number of categories won outright by each team and determines overall win probabilities based on those totals. With the 20 variables, each team wins 10 categories, resulting in an even chance of winning for each of the teams. Using the 23 categories (i.e. including the ratios), Florida Panthers win 13 categories compared to the Edmonton Oilers 10, for a 56 to 43 advantage. So, the 20 variable method results in the series being a toss-up, while the 23 variable model shows the Panthers to be approximately a 2 to 3 favorite (using racetrack tote board figuring). The last rows show the payoff needed on a 10 dollar bet, in order to even the odds. Of course, this doesn't include the government (or bookie, same thing) take-out. So, I wouldn't bet the Oilers as a favorite, but as a moderate underdog, they seem like a good bet. But you shouldn't gamble, this is just for fun.

	Model 1 (20 Var)		Model 2 (23 Var)
Using Category Wins	EDM	FLA	EDM	FLA
Category Wins by Team	10	10	10	13
Probabilities	50%	50%	43.5%	56.5%
Odds	1.00	1.00	0.77	1.30
Racetrack Odds (X to Y)	1 to 1	1 to 1	13 to 8	8 to 13
Payoff needed on a $10 Bet	10.00	10.00	13.00	7.69
Payoff including initial bet	20.00	20.00	23.00	17.69

Model 2 - Using Weighted Scores for Categories

The next model uses the weightings in the data, which are based on win percentages for the better team in each category. This method totals the weighted scores of categories for each team and determines overall win probabilities from those totals. Both the 20 variable and 23 variable models show the series to be very evenly matched, with a slight advantage to Florida. Since the edge in each category is slight, they generally offset each other. Again, the Oilers seem like a good bet if they are underdogs, even if only slight underdogs. Again, however the government will take a large bite, so finding good odds will be difficult, to say the least.

Method 1	Model 1 (20 Var)		Model 2 (23 Var)
Using Weighted Points	EDM	FLA	EDM	FLA
Category Wins by Team	9.9	10.1	11.315	11.685
Probabilities	49.5%	50.5%	49.2%	50.8%
Odds	0.98	1.02	0.97	1.03
Racetrack Odds (X to Y)	98 to 102	102 to 98	97 to 103	103 to 97
Payoff on a $10 Bet	10.20	9.80	10.33	9.68
Payoff including initial bet	20.20	19.80	20.33	19.68

Model 3 - Using Weighted Scores for Categories but Giving all Weighted Points to the Category Winner

This model is a compromise between the other two. It also uses the weighted scores, but gives 0 points to the team with the inferior score in that category. The results are very similar to the first model, though evening up the predictions very slightly.

Method 2	Method 2 (20 Var)		Method 2 (23 Var)
Using Weighted Points	EDM	FLA	EDM	FLA
Category Wins by Team	5.12	5.22	5.12	6.805
Probabilities	49.5%	50.5%	42.9%	57.1%
Odds	0.98	1.02	0.75	1.33
Racetrack Odds (X to Y)	97 to 103	103 to 97	3 to 2	2 to 3
Payoff on a $10 Bet	10.20	9.81	13.29	7.52
Payoff including initial bet	20.20	19.81	23.29	17.52

So, my conclusion based on these admittedly simplistic methods, is that the series should be very close, probably 7 games. Florida has an edge, but Edmonton is only behind by a bit. So, if you see good odds for the Oilers, they would be a good bet, but don't bet, this is just for laughs. Conversely, if you see Florida being the underdog, go for that. But don't really, because gambling is bad.

And here's a picture of the Stanley Cup.

============================================

So now you should read a short story to find out just how bad gambling is. It is only 99 cents, and will be a good investment if it convinces you not to bet on horses or hockey games. :)

A Dark Horse

In “A Dark Horse”, a gambler’s desire to hit a big win seems to lead him to make a Faustian bargain with a supernatural evil. Or is it all just a string of unnaturally good luck?

The story is just $0.99 U.S. (equivalent in other currencies) and about 8000 words. It is also available on Kindle Unlimited and is occasionally on free promotion.