The 2nd round of the 2024 playoffs has been locked into a time capsule for future analysis, as we’re down to the final four (henceforth referred to as R3). This will also double as an R2 review, given how many observations about what just happened should be relevant to the upcoming games. The prevailing theme of round two was “road ice advantage”, which is something we also saw at various points in the 2023 playoffs (Murley commented on Chiclets last playoffs that guys don’t party on the road anymore (excluding certain members of his Colorado Avalanche)). Road teams went 16-9 and if you bet $100 on each of those moneylines, you banked $848, or +1.5 PL $633. That was my best source of success, while my over/under results were awful, at least in the Eastern Conference.
Before we go any further, it’s time for my regularly scheduled obligatory *DISCLAIMER* it needs be noted that I’m not betting with real money. These are all fictional wagers in a spreadsheet. My mission is to engage in a mass betting campaign, picking a winner of every single game, every over/under, because it provides a complete dataset for macroeconomic analysis, which can be shared with you, shedding light on what worked and what failed. I’m also tracking the results of betting every outcome, to help me (and you) uncover previously unknown or newly emerging profit vectors. What started as a thought experiment has evolved into much more.
This might be my first ever R3 Preview, but certainly won’t be the last. Work began on this piece 2 days into the second round, with a commitment publish regardless of my R2 performance. This is my first playoffs sharing picks for every single game, so it’s important to check back in every so often until there’s no games left to bet. In terms of looking back at the previous 3 playoffs for what worked, that strategy was not particularly helpful in the opening round, such that by R2 my eyes were driving many of my picks, not my models. At least you all got to see how my models felt about each playoff game, and I’m sure there were plenty of you who made bets siding with my models, not me. You can admit it, my feelings aren’t hurt. They are my children after all (the song Sweet Child of Mine is on my model building playlist, along with Born to be Alive).
As is the case every May, as the daily slate of game winds down, my summer projects get ramped up. I don’t take vacations or divert my attention to a summer sport, I’m working on one hockey thing or another 365 days per year and total hours per week working on my various projects does not decline when there’s no games on television. Free agency is coming, and that is my primary area of expertise, a passion project on mine that pre-dates betting analysis by decades. I currently have a 120,000-word book manuscript that I’ve been working on since 2020 about the market value of NHL players, but may turn it into a series of blog posts instead. As Bob MacKenzie once said, selling books is way too much work for way too little money.
This gets me to my next programming note; this summer I’ll likely be posting about the NHL draft and free agency. In October, my weekly fantasy and betting reports resume. If you’re only interested in betting and none of the other stuff, you can actually edit your Substack preferences to only get “betting reports”. I’m taking a risk telling you this, because the instructions begin with clicking the “unsubscribe” button in your emails. Don’t worry, it does not immediately end our friendship, but rather takes you to a page where you can actually choose which of my areas of interest, you’re getting notifications for new posts.
Past Playoffs
With each new round brings a smaller historical sample of supporting evidence. Round one had the most data by far, and there was substantial divergence from history. My playoff database goes back to 2021 (2020 was useless), but even 2021 and 2022 were wildly different from 2024, while 2023 fell somewhere in between. The data from 2022 disproportionately deceived my models more than any, when home team boomed to a degree we haven’t seen since. My planned summer model recalibration will not allow any angles to proceed unless they profitable in 2023 and 2024. That pandemic data will never be completely ignored, because it’s still crucial for my Betting Venues model, but the others will be forbidden from making any picks solely on the basis of pandemic profitability.
It’s one thing to review my results from the last 2 playoffs to determine my successes and failures, but I also logged “game notes” explaining my justification for each pick, even when that’s just a gut feeling. Looking back at those notes from 2022 and 2023, my single biggest mistake appears to be the search for “line value”. A majority of my explanations include some reference to line value, or the algorithm being used to measure so-called “value”. If you calculate that a line should be +120, but you’re getting +150, you’re ostensibly getting a +30 bonus taking that wager. It had never occurred to me that value might be bad. What do oddsmakers know that I don’t? Safe to assume they have much smarter people than me on staff.
Fast forward to 2023/24 when I started using/tracking themed betting models; and one of the early models attempted betting based on “line value”. It was so terrible, I had the idea to bet the opposite, and that model kicked ass. It was named “Shorting Value”, and version 2.0 crushed the competition in the 4th quarter of the season. Suddenly the strategy that once dominated my playoff picks (with negative results) was flipped upside down. If every instinct you have is wrong, then the opposite must be right. It’s among the fundamental principles of life. Granted, my aforementioned awesome Shorting Value model collapsed in the current playoffs because it loved pucklines -1.5 goals a little too much.
The first thing that stands out looking at the category results from the previous 3 playoffs, lay-off pucklines -1.5 goals. If you bet $100 on every home team +1.5 AND visitor +1.5 in all 33 of those games, you generated profit from both sides. It might be advisable to put a master-override into all my betting models to lay off pucklines -1.5 goals (at this point it’s probably easier to manually change them instead of re-writing each cerebral cortex). The only team to even produce a profit on the puckline -1.5 goals was Colorado sweeping Edmonton in 2022. Last year Florida swept Carolina and didn’t win a single game by more than a goal.
The trend towards pucklines +1.5 goals began early this season, as they started cashing often in the 2nd round. Fortunately, work on this preview began a few days after round 2 kicked off, and the first investigation was what worked in previous playoffs. Those home teams -1.5 goals were awful in R3 of previous playoffs, and the round that just wrapped. Lay off pucklines -1.5 might be the most important message in this entire report. Granted, it’s entirely plausible that trend will completely reverse, because it is individual match-ups that drive categories, so if there is home dominance in Dallas vs Edmonton, past results get flipped on their heads. We’re playing with very small samples here.
Over/Under
At this stage in the schedule, being an overs bettor gets rough, at least on average looking at previous playoffs (overs went 13-12 in round 2 of 2024). The optimal conditions for overs to thrive are teams struggling at goal prevention, but playoff Darwinism dictates those teams rarely advance to the final four. Bad #1 goalies tend to be golfing come late May, unless the back-up who supplanted them became a hero. Either way, they’re usually not on the ice for conference finals. We already said goodbye to Georgiev, Silovs, and the suddenly struggling Freddy Andersen. We could see a goaltending duel in the Eastern Conference, but high scoring games out west. Below is the over/under table from R3 the previous 2 playoffs.
Thus far in the 2024 playoffs, it’s been a relatively even split between overs and unders, but my own performance was awful in round 2 as my primary advisors did a bad job dispensing advice after a strong opening round. The Eastern Conference did the most damage, taking a big loss on both overs and unders. All my struggles came in the 50% of games where the total was 5.5, which is where it was for the entire Florida-Boston series where my picks went 2-4 (but the 2 winners were minimum bets, and 2 losers were big bets). I’m going to continue sharing over/under picks, but you should consider my poor performance when deciding to tail. If you faded all my R2 OU picks, you had a good round. Maybe that’s the best strategy for R3…
When you’re looking at my pick graphics, the top 5 performing models are the top 5 from the list in the next section, same with ML/PL. I’m regularly checking the best and the worst, and re-calibrating which get top billing. Speaking of models, let’s get to that.
My Model Performance
As mentioned earlier, my eyes have been informing my recent picks more than my models and one of the biggest reasons is lackluster performance. Any accustomed to thriving off pucklines -1.5 goals took another beating, unless they had a predisposition to road favorites -1.5 goals. My “Goalies Last 30 Days 2.0” was by far the best round 2 model, and it was programmed with specific changes for playoffs. This was among the first models built where I verified playoff profitability during construction, which isn’t necessarily a good thing when there’s a shift in what type of picks produce profit. That’s why all the others will special playoff instructions have struggled, but GL30 2.0 found the right angle.
The irony there being that it’s the same goalie starting most games in most series, so it was mentioned in an earlier report that my goalie-based models would be less useful in playoffs. But what’s working is not which goalie I’m predicting to start, it’s the picks it makes using that information that had success. Who I’m picking matters less than the model loving road moneylines, which in the playoffs are quite often underdogs. That was the secret sauce to its success, but we have no idea if those parameters will be transferable going forward. The samples we’re dealing with here are so small, that categories aren’t being driven by league trends, but what’s unfolding in the few remaining series.
Round 2
My results from round 2 were worse on a per series basis than round 1, so that should not inspire heaps of confidence in my round 3 picks yet to come. But at least I’m letting you see exactly where my failures came. Technically I did come out ahead on my series winners, going 3-1, but went 0-4 on my “correct score”. Last playoffs I went 1 for 15 on correct scores, so I need to hit 1 of the next 3 to improve on that outcome. It makes sense that my worst series producing 2/3 of my losses was the one picked incorrectly, Carolina losing to New York. We’ll get to that in my series recap. Up next is the category results from 2024 round two.
The best categories were already discussed in the opening paragraph, and my ugly over/under performance was also already addressed. Perhaps the most alarming result from R2 was home teams -1.5 goals losing $1,700 on $100 wagers (most of those favorites). I did not pick a single puckline -1.5 goals last round, given their crash after the opening game ones in round one. That will continue going forward, as I’m going to continue leaning on Dogs +1.5, at least later in the series (my G1 picks are both ML).
Florida Panthers defeat Boston Bruins
My pick that the Florida Panthers would defeat Boston in 5 games looked foolish when the Bruins blew them out in game one, smarter after game three. We quickly learned Florida had a little rust in game one, while Boston was recently battle tested, which provided an initial spark until the fatigue became a liability and Florida’s fresher legs shook off the rust. My moneyline/puckline picks went 5-1, but was offset by a bad performance on over/under, which was my prevailing theme of round two. The series got nasty, and while Boston had plenty of grit, this was Florida’s mojo all season, battling in the trenches, lighting fires. Boston knew exactly what was coming, but could not beat Florida at their own game.
Once Florida went up 3-1 in the series (my “correct score” pick was Panthers in 5 at +425), I hedged with Bruins +1.5, to eliminate the possibility of losing both wagers. When sharing the pick on Twitter, I made it entirely clear it was JUST a hedge because of the series bet. Panthers dominated games 2 to 4. There was no word on Marchand returning, and nothing aside from Jeremy Swayman gave me any reason to pick Bruins +1.5 goals (aside from underdogs +1.5 goals having a good round two). Boston won the game, but it feels unlikely anyone tailed my pick if they didn’t also have a Florida in 5 bet on the table.
I have watched nearly every game in the playoffs (unless 2 are on concurrently) and not reading much social media. I was surprised to hear on Chiclets that there was controversy that the referees (even the league) were discriminating against Boston and costing them the series. What I saw after the opener was total dominance by Florida in all facets of the game. Boston deserved to be down 3-1. They bounced back strong in game 5 pushing the series to a sixth game on home ice wherein Brad Marchand returned, but that’s where their luck ran out. Sorry Bruins fans, you had no shot at winning the Cup and it’s arguably better to be taken out of your misery now.
New York Rangers defeat Carolina Hurricanes
They say the definition of insanity is doing the same thing over and over again, expecting a different result. Hard to say who’s loonier, me for believing in playoff Carolina again, or Rod Brind’Amour thinking he could beat a team with a Vezina goalie by peppering low-danger shots from the perimeter. He said it in the press conference after falling down 0-3, he said there were two different games on the ice, one they were winning, one they were losing. Part of me genuinely believes that Roddy feels like he’s won the game if they get more shots than their opponent. Shoot more shots, punch your ticket, good job.
That opening paragraph was written after they were down 0-3. Carolina then won game four at home, but I still hit my Rangers +1.5 goals pick. Then the Canes were strangely favored in game 5 in New York down 1-3, so I rolled back that winning NYR +1.5 pick and they lost 4-1. At this point, Carolina in 7 games +500 was my series bet, so if they accomplished the improbable, that $500 would get added to my playoff total (albeit tracked in a completely different spreadsheet). After the series shifted 3-2 going back to Carolina, they were +200 to win the series. Still a longshot, but less impossible than before.
Evidently the two “different games” Roddy was referencing was even strength vs special teams. If Carolina could just stay disciplined and out of the penalty box, they were excellent 5v5. That’s how they won games 4 and 5. For game 6, Rangers opened at +136 on the moneyline, which was also my pick for the game. It was partially a hedge, but the +136 number was very appealing and likely would have been my wager without the series bet. They’ll be facing an arguably tougher challenge in the next round, an opponent with more toughness, better at special teams and more elite finishing talent.
Dallas Stars defeat Colorado Avalanche
The Dallas Stars were my pre-playoff pick to win the Cup and gave me a big scare round one falling down 0-2 to Vegas. They eventually knocked out the defending champions, only to face the previous defending champion, but this time jumped out to a 3-1 series lead, including a pair of wins in Denver where the Avs dominated all year. They got a boost when it was announced that Valeri Nichushkin was suspended 6 months for substance abuse relapse. It would be fascinating to know precisely when he fell off the wagon because if it was a while ago, then wow it was certainly not hurting his output, leading the league in playoff goals. If it happened in Dallas, the league should investigate if there was any “outside assistance” pushing him off the wagon.
The reason for my suspicion is my own joking on this blog during the Jets series that Winnipeg had no chance of coming back unless someone sent hookers and blow to Nichushkin’s hotel room before game five. A week later he was back in rehab. Not sure if I’d be culpable if someone read that on my blog then actually sent hookers and blow to his hotel room. Though I’m skeptical it happened on the road because for sure they would have had a chaperone with him after what happened last season at a Seattle hotel. Even one of the Chiclets podcasts joked they needed someone looking after Val, and they have an exponentially larger reach than me.
Colorado did manage to pick up a big road win in game 5, going back home with their backs against the wall, and Dallas continued their high-altitude dominance with a 2-1 win (high danger chances were 20-8). They outscored the Avalanche 11-3 in their own barn, which is remarkable. Even though Dallas was my pick in game 6 and on the series line, I found myself cheering for Colorado to extend the series because it was so much fun to watch (and Stars in 7 was my series bet), but alas they fell short. The Stars feel like a team of destiny, but we’ll get to their next match soon enough.
Edmonton Oilers defeat Vancouver Canucks
The Vancouver Canucks managed to overcome losing their top 2 goalies to injury defeating the Nashville Predators, but in the process substantially exceeded my expectations, attempting to pounce on a wounded beast. My attempted predation back-fired, making that easily my worst betting series of the first round. I’m not a Canucks fan, but live in Vancouver and am easily lured onto their bandwagon, especially in playoffs when the sirens sing. Sorry, that was a reference to Greek mythology, not the sound of first responders downtown when Vancouver loses in those same playoffs. There was no way I was going to pick Vancouver and their 3rd string goalie to beat the mighty-ish Oilers.
Oilers were my series pick, but taking them in 5 games was a loser after game three when they fell down 2-1. The situation was bad enough that starting goalie Stuart Skinner was benched and the back-up (who I jokingly referred to as Captain Pickard earlier in the season before he proved himself decent and the jokes ceased). Ostensibly they were copying the Vancouver strategy of running with the substitutes, which worked, tying the series 2-2 heading back to Vancouver. All of a sudden, the Vancouver back-up was getting outdueled by his doppelganger at the other end. The Canucks convincing victory in game five lured me onto their bandwagon, winning the high danger scoring chances 18-10.
Whatever line matching strategy Tocchet used to turn up the attack in game five completely vanished in the next 2 games, as that bandwagon broke down at the side of the road shortly after my boarding. My pick was Vancouver for game 7, before finding out that Brock Boeser would be out with blood clotting issues. He was key to their previous playoff success and was a devastating loss. The Canucks managed to score some late goals to make it close, but ultimately didn’t have enough firepower to overcome Edmonton.
Round 3
Florida Panthers at New York Rangers
The first confirmed R3 match-up features the top two seeds in the Eastern Conference, which according to my spreadsheet is only the second time since 2014 we’ve seen the division winners face-off in round 3 (the other being Tampa vs Washington in 2018 (which also means the Pacific and Central winners have not met in the conference final since at least 2013)). The Rangers won home ice advantage by being 4 PTS better than Florida in the regular season, but the Panthers had a +68 goal differential compared to Rangers +53. That’s compelling evidence Florida was a better team. Yes, they start on the road but did go 4-1 on the road in the first 2 rounds.
Since my bracket had neither of these teams advancing this far, I’m not emotionally invested in either option prevailing. In my defense, picking Toronto to beat Florida in my bracket was because I hate the Leafs and did not want that outcome happening. That may not be an entirely “defensible” position for those of you here seeking betting advice, but if you read my justification for picking Toronto to beat Boston and that convinced you that Toronto was going to win the series, that’s on you. Reverse jinxes don’t work if you openly state that’s what you’re doing. My series bet is Panthers -140, which has moved to -150, but do think it will be close and am taking them to win in 7 games at +500 (I’m only 1-11 on correct score bets in playoffs, but a few came close).
All that being said, there is probably more value on the Rangers at +130 given they enter this match-up with the hotter goalie, but we’ve seen flashes of elite Bobrovsky at some key moments in earlier rounds. Bob could catch fire at any moment. What’s interesting is that Florida is a -150 favorite, but Rangers in 7 is +475 while Panthers in 7 is +500. So oddsmakers think the Rangers will be favored if this goes 7 games, which does suggest the series line might be off. Given that Florida in 7 is my bet, I’m really hoping the math Draft Kings used to make that game 7 assessment is the part that’s out alignment.
Game One
The Panthers lost game 1 of the previous round while the Rangers won, but Florida had rust and Boston was fresh. Both these teams won their previous series in 6, so the rusty vs fresh dynamic should not factor (aside from the Rangers proving last round they were good with rust). 63% of my model money is on Panthers moneyline, including a big wager by Betting Venues. The bad news is that my best performing model has the Rangers ML, so this is not as obvious a pick as the 63% suggests. My best performing playoff models preference for Florida ML is what I’m tailing. I also really like over 5.5, mostly because there could be a ton of power plays in this one, especially if Rempe suits up and sets off fireworks. My best models/algorithms liked the under though, so it’s by no means a lock (yet I’m still making a max bet hoping July 4 comes early). Also recall my terrible eastern OU output in R2.
Edmonton Oilers at Dallas Stars
It needs to be disclaimed at the onset of this discussion that I’ll be cheering hard for the Canadian team to win this match-up, which risks injecting some bias into my pick selection. Fortunately, my betting models have no such patriotic affiliation and their preferences for every game will be shared on social media. Dallas was my pre-playoff pick to win the Cup and I’m sticking to my guns. That’s not simply a reverse jinx attempt, I genuinely believe Dallas will win this convincingly, having just ousted the 2 most recent Stanley Cup champions. I’m cheering for them to lose but believe they’ll win. They have the best depth, and the far better goalie.
My pick for the series was made at the 10-minute mark of 2nd period, game 7 with Oilers up 2-0 because Draft Kings already had the series line posted. With the game at the halfway marker down 2, that’s all Draft Kings needed. Granted, that line could have been up far earlier than that, that’s just when my eyes confirmed it. I immediately went to Twitter and posted my pick, Stars to win the series at -115 and in exactly 6 games at +550. Dallas sweep at +1200 was tempting, as hitting that would erase my deficit on series and “exact score” bets in the playoffs, but didn’t come to that epiphany before Tweeting, so I’ll stick with the original.
My heart wants the Oilers, but the team is far too top heavy to deal with the Dallas depth. Even if they can win the McDavid-Draisaitl minutes, they are going to get destroyed when they’re not on the ice. Not to mention the massive mismatch in goal, with Jake Oettinger sporting a .918 playoff save percentage while Stuart Skinner is at .881. My concern is Oettinger starting 18 consecutive games, where heat can be an issue in Dallas home games. We’ve seen him run out of gas in past playoffs. The Oilers have 2 excellent defensemen, but Nurse and Ceci are going to be a massive liability against the Stars attack. Being proven wrong would actually make me happy in this case, it just feels entirely unlikely.
Game One
My largest model position is Dallas moneyline, which is going to be my pick too. Granted, Stars did lose the opening game on home ice the previous 2 series, and there might be a little rust to shake off from their wait (which should benefit Oettinger the most, followed by Heiskanen who logs the heaviest minutes). The rust factor is why I’m only logging a minimum wager. My best models are divided. The 2 playoff leaders among my regular season models have Edmonton moneyline, while my two hottest playoff models have max bets on Dallas (one ML, one -1.5 PL). There is a case for both sides, but since Tweeting my intentions to bet Dallas game 1 midway through EDM-VAN game 7 has me locked into this position.
The over/under is an even tougher decision. With my top models/algorithms taking over 6, but my Small Council and playoff models preferring under. That potential rust on Oettinger (who let’s be honest also needed some rest) is one reason I’m going to side with my best performers and take the over. The other reason is overs going 4-0 in game 1 of the previous round. Granted, there may not be as many power plays in game 1 of this series (compared to the potential fireworks in New York). Edmonton isn’t exactly a dirty team, and Dallas will certainly be trying their best to stay out of the penalty box. We could see both teams on their best behavior trying to avoid playing shorthanded.
This may be my last betting report of the season. Good luck!