Create Account

What Are the Odds - A Playoff Analysis
#1

As we enter week 4 of S80 and are ramping up for playoffs, I thought I’d release this little article.
At the start of the S79 SHL playoffs, I found myself filling in my playoff predictions and having absolutely no clue what to think. At the time, I was entering my second playoff period in the sim league and knew very little about the big boys and girls league. My limited knowledge stemmed from the few interviews I had with SHL GMs and Scouts before the S79 SHL Entry Draft. As I sat looking at the standings and learning about the existence of some of the teams for the first time I had flashbacks to my less than successful ability to predict the NCAA’s March Madness bracket. Another tournament that I know nothing about. However, this time I couldn’t base it off of the coolest mascot or the team with the funniest name. In sports, and even more so in March Madness, there always seems to be some big upsets that happen for the first time. But I was lost when it came to figuring out which upsets could occur in the SHL. So I did what any insane person, who clearly has too much time on their hands, would do. I calculated the historical odds of every match up in SHL history based on conference seeding.
Now, I started this project with the intent of getting my much needed insider information before the S79 playoffs predictions were due but soon came to the realization that the index only contained S53 onward. I began to dread the enormous task of gathering all the data required to put something like this together. Little did I know that @luke and his team were almost ready to publish the historical data that I desired. Timing could not have been better! So thank you to everyone that put in all that effort to load the historical data onto the index for us schmucks who like to nerd out.
Before I get into the data portion of this, I want to emphasise a few things.
1. I’m just some guy who threw a butt ton of data into an excel spreadsheet with some formulas. I can almost guarantee there will be mistakes. 
2. The league has gone through great growth and numerous playoff format changes which created some weird conference matchups. The changes in the playoff formats over time mean that conference matchups weren't always the best analysis of matchup rankings, but I stuck with it to be consistent.
3. I understand that more goes into winning a series than the conference rank of the team. Team Tactics, GMing, etc. all have a major role in who is victorious but I wanted to be able to look at the matchups without considering who was on the team or who was coaching them.
First and foremost, we will look at the history of the SHL. There has been 721 playoff series in the SHL since the inaugural season. That is a lot of playoff hockey! Of those 721 series, 62.67% (449) ended in an expected result (higher conference seed won the series). That leaves 34.12% (246) for upsets and 3.61% (26) as even matchups (e.g. 1 v 1, 3 v 3, etc.). It’s crazy to me that almost a perfect two thirds of the playoff matchups have gone as expected.  Let’s break it down further.
There are 28 possible matchups that we can look at. Seed 1 v 8, 1 v 7, 1 v 6, 2 v 5, 2 v 7, 5 v 6, etc. The list goes on for 28 pairings. Just trust me…. Or don’t…. I’m not a mathematician. Since season 1, we have seen at least one playoff series in all but 3 possible matchups (Figure 1). We have never seen a 5 seed play against an 8 seed, a 6 play 8, or a 7 play 8. Many of the trends we see in the historical data can be attributed to the playoffs changing size over time. This is one of those cases. It wasn’t until S58 when we saw an 8 seed for the first time. I suspect with time, we may eventually see these matchups. You can see that 1 v 4, 1 v 3, 1 v 2 and 2 v 3, are by far the most common matchups. Back in the early days (S1 & S2) three teams from each conference made the playoffs. From S3 to S45 we saw 4 teams from each conference make playoffs, making up the bulk of these matchups.
[Image: Screenshot-2024-12-31-09-38-08.png]
Figure 1 - The number of times a matchup of two specific seeds has occurred. I included graphs just for you Luke. Keep in mind these numbers may be more than the total # of seasons in the SHL as the same matchup can occur more than once in a playoffs.
Now, I’m going to explore the rare matchups a little more. Partially because I’m interested, partially because you're interested (ha!) and partially because it gets me more money. I was planning on using <5% as my definition of rare. That would mean exploring any matchup that has occurred 36 times or less. Nineteen of the 28 matchups would be considered rare and that’s just too much for me. So I’m going “extremely rare” with a definition of occurring <1% of the time, meaning 7 or less times in the history of the SHL. Funny enough, 7 matchups have occurred 7 or less times. They are: 1 v 7, 2 v 8, 3 v 8, 4 v 7, 4 v 8, 5 v 7 and 6 v 7. Again, seeing these lower seeds (e.g. 6, 7, 8) shouldn’t surprise anyone given their relatively recent inclusion in the playoffs. 1 v 7 has occurred three times with the only upset coming in S74 when the San Francisco Pride beat the Edmonton Blizzard 4-2 in round 2. That wasn’t the only record the Pride set during the S74 playoffs as they beat the Los Angeles Panthers (seed 4) in the first round. The 4 v 7 matchup has occurred four times with all but the one described above going the way you’d expect. Although their run stopped after the second round, back-to-back historical feats is very impressive. 2 v 8, 3 v 8 and 4 v 8 have all only occurred once with the higher ranked team winning each series. 5 v 7 has occurred five times with three expected results and two upsets. The two upsets occurred in S78 and S79 where we saw Buffalo (7) beat Toronto (5) to make history followed by Minnisota (7) beating New Orleans (5). Last but not least, the 6 v 7 matchup. This matchup has occurred four times in the SHL. Three of the four have ended with the 6 seed winning but in S65 we saw the the Hamilton Steelhawks (7) beat the Toronto North Stars (6) 4-2 in the first round.
Now, I wanted to move onto teams that were on the wrong side (or right side) of history, upsets that occurred for the first time recently. I’ve described a few above but I’ll go over them again. We’ll look at all upsets that occurred for the first time in SHL history since the last expansion prior to S60.  Only two 1 v 6 upsets have occurred, in S67 and S68. In 67, we saw Montreal (6) beat Toronto (1) and in 68 we had Manhattan (6) beat Atlanta (1). The 1 v 7 upset has only occurred once, in S74 when San Francisco (7) beat Edmonton (1) in the second round 4-2. Moving onto the upsets occurring against the second seed. The number five seed has upset number two, four times. The first time was in S63 when Los Angeles (5) beat Chicago (2) in the conference finals 4-1. A 2 seed has upset the 6 seed 3 times, occurring first in S57 (slightly outside our defined range) and most recently in S79. In S57 Calgary (2) lost to Atlanta (6) in a cross conference first round matchup (.....weird playoff format but okay). Most recently, Atlanta (6) beat New England (2) in the first round 4-2 in S79. Onto upsets involving the 3 seed… The third seed has lost to the sixth seed four times. The first occurred in S68 with Buffalo (3) losing to Manhattan (6)  4-1 in the conference finals. The 3 seed has only been upset one other time recently and that was to the 7 seed in S79. Toronto (3) lost to Buffalo (7) 4-2 in the first round.  Only three more matchups have had upsets occur since S60. The fourth seed has lost to the seventh seed once, occurring in S74. Los Angeles (4) lost 4-2 to San Francisco (7) in round 1. This is the only time this upset has occurred. Seed 5 has lost to 7 twice, occurring in S78 and S79. The upsets are described above but I’ll refresh your memory. Buffalo (7) beat Toronto (5) in S78 followed by Minnisota (7) beating New Orleans (5) in S79. The final upset matchup is seed 6 beating 7. I know, hardly an upset. Occurring once in SHL history in S65 when Toronto (6) lost to Hamilton (7) in the first round.
Maybe you’ve noticed the same trend I have or maybe it’s my real life Canadian and anti-Leafs bias coming in but I have to say, it feels like Toronto is on the losing end of many of these historic upsets. To me, that feels like the most Toronto thing ever. Who knew real life Toronto-isms would play a role in our sim league. I didn’t officially look at how many teams or franchises were on the winning or losing side of the historic moments but figured I’d throw Toronto under the bus (sorry, not sorry). Toronto was the first 5 seed to lose to 7 (1 of 2 matchup upsets), first 6 seed to lose to 7 (only matchup upset), first 1 seed lost to 6 (1 of 2 upsets), side 3 seed to lose to 7 (only matchup upset).
[Image: Screenshot-2024-12-31-09-43-04.png]
Figure 2 - Lawlz Leafs
If you’ve made it this far, you are a trooper. For that, I thank you.
I know you’ve been thinking this whole time that you don’t really care about what happened for the first 60ish seasons of the league and you probably feel those don’t really count because of how different the league is now. Well guess what! I had the same thought. So grab a snack and buckle up cause we’re diving into more recent times to see if anything is different. This is a look at the playoffs since the transition to FHM8 or S66.
Since switching to FHM8, we have seen 56 upsets out of 210 series, that’s 26.67% which is just under 10% lower than the historical upset percentage (34.67%). The top four most common upsets in this period of time are: 1 v 4 (6/14 upsets; 42.86%), 2 v 3 (6/8 upsets; 75%), 3 v 5 (5-13 upsets; 38.46%) and 3 v 6 (4/10 upsets; 40%). It’s interesting to see how the number 3 seed is the most likely to upset.
When comparing FHM8 to the historical data, you can see the percentage of upsets in each matchup is remarkably similar to the historical data (Figure 3). We can see that the largest difference comes from the 2 v 3 matchup. Historically, this upset occurs 46.88% of the time while it has occurred 75% of the time since S66. I expect this will return to normal as we continue to put more seasons between us and the transition. The other significant difference (using my anecdotal measurement of significant, not statistically significant) is between the 1 v 2 matchups. Historically, we see this matchup go either way almost 50% of the time (48.81% upset, to be exact) but recently we’ve seen the number of upsets be much less at 18.75%. Either the number 1 seeds have gotten much stronger in FHM8 or something else is up. We may be able to attribute this to the sheer number of matchups that have occurred between the 1 and 2 seed teams historically compared to recent times. Historically, the 1 v 2 matchup has occurred 85 times (third most of all matchups), whereas since S66, it has occurred 16 times. Give it another 52 seasons and we might see this matchup become more even.
[Image: Screenshot-2024-12-31-09-38-38.png]
Figure 3 - The percentage of matchups that end in an upset.
I was hoping to get more out of the FHM8 data, but lucky for you, I didn’t.
Moving right along, I also analyzed the seeding of the cup champion for all 79 seasons. Due to an seemingly ever changing playoff format the data may be a little funky but in the end, I looked at the divisional, conference and league seeding of the cup winning team. This stemmed from the idea of trying to predict upsets and seeing how far a team would make it. Is a team going to win the cup as a 6 ranked team in the league? I had no freaking clue so I crunched the numbers. To answer my own question, yes, it is possible albeit unlikely.
Thankfully, the conferences in the SHL have remained the same since the dinosaurs roamed the earth. Looking at you SMHJL, which seems to switch conference names every 20 seasons. Also why was Newfoundland in the Pacific division in S70 when Newfoundland could not be more in the Atlantic?!? The island of Newfoundland and the shores of Labrador are surrounded by the Atlantic ocean.  Anyway….. That’s a discussion for another day. In the SHL, the West has won the cup 43/79 times or  54.43%. That’s pretty even which is awesome to see. Recently, the cup seems to bounce back and forth between conferences relatively often. We haven’t seen a conference go on a run of more than 3 cups in a row since the late S20s when the west won 9 in a row.
Divisions were added into the SHL in S47 with the most recent division set up coming with the expansion in S60. Since the new divisions, the Pacific Division (Calgary, Los Angeles (3 cups), Edmonton (1), Seattle (2) and San Francisco (2)) have won 8 out of 20 seasons or 40% of the time. A big part of this can be attributed to Los Angeles and San Francisco winning four out of the last five cups. Next, we have the North East Division (Buffalo (2 cups), Hamilton (2), Toronto, New England (3) and Montreal) who have won 7 championships in 20 season (35%). The Central Division (Chicago (1 cup), Texas (1), Minnisota, Winnipeg (1) and New Orleans) is next at 3 championships (15%). In last, we have the Atlantic Division (Baltimore, Atlanta (1 cup), Tampa Bay, Philadelphia (1), Manhattan) who have won 2 cups or 10%.
Table 1 - Divisional Championship percentages
[Image: Screenshot-2024-12-31-09-45-23.png]
When considering your playoff predictions, you can use the following table to determine if a team has a chance to win. There are a number of League, Conference and Divisional ranks that make the playoffs but have not won a cup. So, although possible, history has shown it is not going to happen.
Let’s explore those “rare” champions. This time, I’m going with the <5% definition of rare. That is any of the ranks that have won the cup less than 5% of the time. That would be League rank 8-10, Conference rank 5 and Divisional rank 3. Starting with the rare League rankings, the 8 seed has won the cup once, in S22 when the Manhattan Rage lost more than they won in the regular season with a record of 25-22-5. The Rage beat the league seed 2, 3 and 7 on their way to win the cup 4-3 over Los Angeles. The number 9 seed has won the cup twice, both by the same team and relatively recently. The New England Wolfpack managed to win the cup as the 9th seed in S74 and S77. In S74, they beat Hamilton, Buffalo, Philedpehlia and Winnipeg. Interestingly, Winnipeg was 10th in the league so it was a 9 v 10 matchup in the finals. The S74 victory for the Wolfpack was also the only time a 5th seed in Conference rankings has won the cup. In S77, they beat Toronto, Buffalo, Philadelphia and Texas. As a side note, anecdotally, the New England Wolfpack seem to be at the forefront of upset history a lot of the time. I don’t have any data to back up that claim but while analyzing the data, they are one team that stuck out to me as a team winning as a lower ranked team. Lastly, the 10th ranked team in the league has won the cup once. In S43, the West Kendall Platoon beat Hamilton, Buffalo and Winnipeg to win the cup. As for the final ranking “rare” feat, in S71 we witnessed Buffalo make a historical championship run as the only 3rd ranked divisional team to win a cup. During this run, Buffalo was ranked 5th in the league and 4th in their conference. On their way to winning the cup, they beat Montreal, New England, Philadelphia and Winnipeg.
Unsurprisingly, we see a bunch of lower ranks that have never won the cup. Given, the changing playoff formats, I have no clue what the math is for which ranks have never had a chance at making the playoffs. All I know is it is possible for a 5th or higher ranked Divisional team, an 8th or higher Conference ranked team and a 17th or higher League ranked team to make the playoffs, which means it is possible for those teams to win the cup. Yet many haven’t.
Table 2 - The percentage each rank in the league has won the cup.
[Image: Capture.png]
That will wrap up the SHL playoff ranking analysis. I doubt anyone actually made it through all that jiberish but if you did. Thank you for coming along for the ride. Hopefully the errors are few and far between. I have collected most of but not all of the SMJHL data and will be doing a similar analysis one day. My intent was to combine the two but the SMJHL is more complicated as the format changes more times than my underwear. There are a number of factors that go into which team wins but maybe this will give you an additional tool to look at when submitting your playoff predictions.
Go Kraken Kraken  & Go Pats!  Patriotes
Reply
#2

ghamss my goat

[Image: 1161116581909246052.gif?%60] [Image: kahri.gif][Image: 1161116581909246052.gif?%60]
Reply
#3

absolutely love that kind of media!

[Image: KSelich.gif]
Thank you all for the amazing sigs & player cards
Germany Citadelles  Stampede [Image: vhY18i8.png][Image: Raptors.png][Image: gs89eGV.png] [Image: eE2UQZC.png] Stampede Citadelles Germany



3. Buffalo Stampede , Eduard Selich 5 (Maximilian Wachter, Alexis Metzler) at 16:25
5. Buffalo Stampede , Eduard Selich 6 (Steven Stamkos Jr., Brynjar Tusk) at 19:48
8. Buffalo Stampede , Eduard Selich 7 (Brynjar Tusk, Alexis Metzler) at 13:55
9. Buffalo Stampede , Eduard Selich 8 (Anton Fedorov, Mikelis Grundmanis) at 15:12
10. Buffalo Stampede , Eduard Selich 9 (Dickie Pecker) at 19:43 (Empty Net)
Reply
#4

amazing media Ghamss love to see it.

I think its also funny to note that there have only been 2 reverse sweeps in the finals of the SHL. S29 and S50. Where Calgary reverse swept Hamilton(S29) and Tampa Bay(S50)

[Image: 0XJkcN5.png]
Czechoslovakia PROFILE || UPDATE || RAGE. Rage 
[Image: luketd.gif]




Reply
#5

quality media! lets go!

[Image: yXL2mHE.png][Image: 9xZZrYI.png][Image: vshUpkM.png][Image: B7juH4M.png]
Reply
#6

I cant believe GHamms would be so mean to TOR what have we ever done to you :eocry:

Its not like I've told Kahri to dungeon you several times.

[Image: pyro182.gif]

Thank you hewasajazzman, Slothfacekilla, Sulo, Frenchie, and Raymond for the sigs Smile
Reply
#7

01-08-2025, 02:39 PMPyro182 Wrote: I cant believe GHamms would be so mean to TOR what have we ever done to you :eocry:

Its not like I've told Kahri to dungeon you several times.

It's my IRL bias coming into it.

I don't believe the dungeon actually exists! I've been threatened many times but never put in there.
Reply




Users browsing this thread:
1 Guest(s)




Navigation

 

Extra Menu

 

About us

The Simulation Hockey League is a free online forums based sim league where you create your own fantasy hockey player. Join today and create your player, become a GM, get drafted, sign contracts, make trades and compete against hundreds of players from around the world.