NHL Playoff Bets Friday April 26

Welcome to another picks post, the first since Monday as more off-days were mixed into the scheduling. There was a picks email exclusively for free subscribers on Tuesday (for the Wednesday games), where I jumped on the home team bandwagon (in my defense, they went 7-1 in game ones) then road teams went 3-0. The 2 previous playoffs, road teams were fantastic in game one, bad in game two. I actually speculated that might flip because losing game one increases the pressure to win game two (the essence of “zig zag” betting) but still bet all the home teams. At least Dallas and Edmonton also received glowing endorsements the next day on Spittin’ Chiclets, so my enthusiasm was not without merit (unless Chiclets is without merit).

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.

One note on the pick graphics below, there was some performance-based shuffling of the models contributing. We are down to 7 “Cupwraiths” as the worst of the new advisors have been dropped from the display, which may require a re-branding from “the nine” (yes there were 7 rings of power gifted to the Dwarves in Lord of the Rings, but less notorious). The zig zag models probably should have been dropped too, but they were just terrible in game one (when nobody lost the previous game) and have improved since real zagging became an option. The Game Sum 3.0 model began 8-2 then went 0-8, mostly picking home teams -1.5 goals, which started off great, then crashed (that also brought down Game Sum 2.0 who also started strong on the backs of H-1.5).

The so-called “Expected Goals Value Voodoo”, named because I had no idea why it worked, turned out it does not work (missing on 11 consecutive picks). Just because you can see a pattern in a small sample of numbers, doesn’t mean it’s predictive when applied to new games (add that to my model building commandments). xGVV will continue making picks, but you won’t see any of them reflected in the pick graphics. Also, those 9 new playoff models had a disproportionate impact on “model % invested” because there were twice as many as the other aggregators. That stat was changed to just be an average of the 3 groups (but still summing to exactly 100% each game (“rounding error” notwithstanding).

The Profit Aggregator has returned to replaced “Largest Model Position” underneath my own pick and is performing well in the playoffs. After the last 2 days, I’m now down $1,000 pretend dollars with my own picks (I was above the profit line 3 days ago). We’ll find out soon enough if that indicates an emerging cold streak, or just bad puck luck. I’m hoping shuffling out the worst performing models (and shuffling in some previously demoted who are actually having a good playoffs (like Betting Venues suddenly got hot, Tailing History 3.0 went cold and was demoted)) improves the quality of the picks you’re about to read. So, let’s get to those, shall we? (parlays at the bottom)

One final note of low significance, the “Macro Daddy” who started 0-10 spawning the creation of Daddy Macro who bets in reverse, went 5-0 after that. Neither Kris nor Kross is showing up in the pick graphics though (probably less than 15% of you are the right age for Kris Kross jokes, it’s a very niche reference). Actually, after writing the above note, I found an error in the Macro Daddy cerebral cortex which led it to picking visitors nearly every game (not affecting any other model). That has been corrected, so we’ll see if the model improves.

New York Rangers at Washington Capitals

The Washington Capitals are returning home down 0-2 to the beast of the east, who let us down last game by failing to score an empty net goal (and/or prevent the 3rd Caps goal) but still won. Rangers -1.5 goals was very popular in parlay polling. Washington opened as a +160 dog, which initially looked appealing (that has moved +142, so the public found that number appealing too). Still, a large majority of my model money is on New York, and they are getting destroyed on Washington HMLs in the last 30 days (and Rangers V-1.5 for that matter). That’s why I’m taking the road moneyline at -192 (my pick was recorded at -192 and I’m not changing it to -170 just to get a higher return, that’s against the rules).

Over/under is more disputed. My Small Council wants under 5.5 goals, but my best performing advisors (including the still hot Expected Goals Last 30 Days version 2.0) like the over, which I’m going to tail. My confidence in that pick is maybe a 4 out of 10, but abstaining is also a violation of my mandate.

Vancouver Canucks at Nashville Predators

My pick for game two of this series was made before finding out there was anything wrong with Thatcher Demko and suffered a big loss on that game (DeSmith should have helped the over, but Vancouver couldn’t score enough goals). Demko was a big reason why I liked Vancouver in that game, but now my confidence in this team has dropped significantly (by the way, I Tweeted before the game they were still +210 on the series line, but that dropped to -105 after the game, hope some of you hit that before the drop). My series pick is Predators +130, so I’m feeling good about that one (though not enough to overcome my despair having Dallas as Cup champ, but we’ll address that tomorrow).

Casey DeSmith is the confirmed game three starter, and my betting models feel very strongly that Predators -1.5 goals is the best bet on the board, so I’m tailing at +190 (which has moved to +205 with the Canucks taking some public money today). The public and I do not see eye to eye on this one. No doubt the Canucks are under tremendous pressure to pick up a win in Nashville, and I’m still cheering for them to do that, just don’t think it’s very likely unless they have a goalie with the ability to steal games. My Small Council is laying a max bet on over 5.5 goals (which missed last game), but xGL30 2.0 has the under. Granted the expected goal models don’t care about goalie impacts. I’ll tail with the over because of DeSmith, but a slightly smaller wager.

Winnipeg Jets at Colorado Avalanche

The Colorado Avalanche picked up a huge win in game two, returning home with a split, now a -160 favorite to win the series (Avs -130 is my series pick (in 7 games at +500)). Game two of Jets-Avs was my single largest return in the playoffs thus far, with a max bet on over 6 goals and a smaller wager on the Colorado moneyline. Although my justification for Avs ML was lacking in confidence, mostly due to concerns about Georgiev being awful in the previous game (with equally awful body language). It makes sense why Avs ML received little support in parlay polling, leading to Jets ML being thrown into the final.

The Jets actually attracted 52% of my model money, especially -1.5 goals (which they hit in 8 of their last 13 road games). Looking at my model returns last 30 days on Winnipeg road games and Colorado home games, the home side is a big loser, the road side is a big winner. There is data supporting the Jets, and I’m cheering for them to win, but just don’t feel like they have much of a chance to win this series. I’d love to be proven wrong, but I’m probably betting Colorado for the rest of this series. At some point Hellebuyck will probably steal a game, but his performance in the first two left me lacking confidence that’s happening. Colorado ML -166 is my pick (that’s moved to -175), also really like over 6.5. My models largest position is Avs -1.5 goals, but I’m sticking to the ML.

Edmonton Oilers at Los Angeles Kings

The Los Angeles Kings stunned the Spittin’ Chiclets crew and pulled off an improbable game two series victory, returning home all tied up. The Oilers offense continued rolling, but goal prevention was insufficient (at least I’m 2 for 2 on overs in this series). My largest model position is Edmonton -1.5 goals, they just have a bad record betting that in the last 30 days (though only on a small number of wagers because Oilers only played 5 road games in that span). Picking over 6 goals was the easy part (that has moved to 6.5, and I’m guessing most of my models would still prefer over (pause to check) the Small Council flips to under at 6.5, but the other 2 aggregators (and xGL30 2.0) still really like the over.

This was probably the game that took me the longest to make-up my mind, which comes after confidently taking Edmonton in game two. Most of my best performing models are only logging small wagers on this one, with 27% on Kings -1.5 goals. That was not an option I was willing to take, but Kings ML +110 as a home dog was very tempting (my only model taking VML is Shorting Value (who is my best model) with only a minimum wager). That aforementioned Oilers -1.5 goals bet is currently at +185, so maybe I’ll just lock into the puckline with a minimum wager. Small risk for a larger reward.

Parlay Combinations

My playoff parlays are currently 0-9, although most of those were democratically elected, so the blame doesn’t fall squarely on my shoulders. This puts my career parlay return below zero on even sized wagers, at least when you include playoffs. My regular season parlays finished about the profit line. Honestly, it’s just so difficult to pick winners in playoffs that perhaps trying to build 3-leg parlays is a futile exercise. In 5 elected parlays, even the two highest vote recipients both hit only once in the same parlay, going 1-4 if this was entirely a 2-leg endeavor. I’m up to 67 career parlays, so perhaps I’ll do 2 more and call it a wrap.

That’s a joke, of course I’m not stopping at 69. By the way, I still have no idea why those finishing 3rd and 4th in polling are substantially better bets than 1 and 2. That’s also the same when I’m ranking them in the absence of democracy. One of the reasons I still have no idea is because that’s on the summer checklist with no investigation having taken place. Parlays are getting shut down entirely soon enough, right now I’m just building up a sample size. That’s why you should only be making small bets on these as longshots.

My Picks

Rangers ML, Colorado ML, Nashville -1.5 PL, which currently parlay to +674. Add Edmonton -1.5 PL for +2105.

Largest Model Positions

Nashville -1.5 PL, Edmonton -1.5 PL, Rangers ML, which currently parlay to +1303. Add Colorado -1.5 for +3338.

Sorry, 1303 was the year Edward Longshanks took back Scotland from Mel Gibson William Wallace, so I don’t think I can endorse this one unless one of the lines moves (I am of Scottish ancestry), but if you’re British, by all means hammer that one. The 4-leg feels fun for 33 to 1.

Pure Moneyline Zig Zag

If you parlay the losers of the previous game, here ranked by the % my models wagered on that team for this game.

Vancouver ML, Winnipeg ML, Washington ML, which currently parlay to +1058. Add Edmonton ML for +1948 (a lucky number if I have any Israeli subscribers).

I’ll be back tomorrow. My picks for Thursday were shared on Twitter, so you can see those in my history. Tampa ML, Carolina ML.