Locky’s First-Period NHL Over/Unders: 5 Games My Model Likes on Tuesday

Good morning! (Or good … whenever you're reading this.)

Today at The Action Network, I've decided to start a project. A series, if you will.

I've been highly successful in a limited sample size betting first-period NHL totals, and as a market, they've become more a part of the discussion among bettors in general.

So instead of putting my projections on Twitter as a generic, bland screenshot, we're going to do better and house them in an article.

There is absolutely no guarantee these will succeed. But they have succeeded so far, and whether it leads to praise, criticism or scorn, sharing them is something I'd like to do.

We might as well all figure out together whether they're any good or not. Since Jan. 1, they have produced a 23% ROI on plays with a 5% or greater edge, escalating upwards as the edge increases. Even at low edge amounts (2-5%) the ROI is about 9%.

So since those have had SOME value thus far, 2% is the threshold we are using to justify a wager.

Here's what you can expect: Each day (except when the slate is too small) I'll do a brief write-up of my projections here. But most importantly, I'll also be providing an Excel document with my projections at the bottom of each article.

Why is that important?

Because you may be reading this at noon, or at 6 p.m. ET, or right before puck-drop of the last game of  the night.

By giving you the freedom to input your own prices from your own sportsbook, you can see if there's any edge remaining — or if the edge has actually increased.

This is way more valuable than me saying "Stars Under -110 an easy winner!" and you say "Wait, it's -140 now, what do I do?!" Those types of interactions are incredibly uninteresting and don't solve any problems.

Our bet sizes should be different if the prices are different.

Anyway, some thoughts on today's projections, which are based on openers as of 10:30 a.m. ET, Tuesday.

Large edges everywhere!

Today is the perfect storm, because we were going to launch this anyway, and it just so happens that, based on the openers, there are massive perceived edges in the market. The last two days … nothing.

But of course, when we're doing this project … we have high percentages in five games! Here's some more analysis on three of those games. The full breakdown of the slate is at the bottom of this post, along with the aforementioned link to the Excel doc.

Dallas Stars at Florida Panthers

7 p.m. ET

Dallas Stars Unders are frequently a high-edge play, and for good reason. As a team, the Stars have the profile of the most conservative team in the NHL at the start of the game. They don't score first-period goals or allow them with frequency. Only 18.2% of Dallas' total goals scored (excluding overtime and shootout) have come in the first period. The league average hovers around 30%. Wow.

Pick: Under -105

Washington Capitals at Columbus Blue Jackets

7 p.m. ET

The Capitals are always a popular "over" team because of their star power, and last night was no different against the Kings, as that game flew over every derivative total in every way.

Tonight the model thinks that Washington, on a back-to-back against Columbus, has value with that game's first-period over as well.

Pick: Over -155

Calgary Flames at Tampa Bay Lightning

7:30 p.m. ET

Tampa Bay gained some headlines (along with Chicago) for going over their first-period total an absurd percentage of the time, but that doesn't mean you just blindly bet them every night.

Some nights, there isn't an edge because the juice is crazy (-180, -190, or higher). Just because a team has gone over a lot, that isn't necessarily predictive of how they will do in a match-up against a specific opponent.

Anyway, the edges, as well as the document are linked below. For an FAQ of just what exactly we are doing with this series, that can hopefully be even more helpful, click here.

Pick: Over -175

Locky's First-Period NHL Over/Unders: Full Slate Breakdown

Download the Excel doc to input odds from your sportsbooks. The table below is best viewed on a desktop computer.