I hope everyone is enjoying their holiday season. After taking last week off we are back in action with a few more articles to help prep for drafts. In the last couple we have taken a look at some stat measures (here and here) that can help us understand projections or even inform our own, but this week we are going to turn toward some projections already widely available. As much as you should be buying the Dobber Fantasy Guide, today is not about the guide, but about Yahoo.

Yahoo is important because a huge number of leagues use Yahoo. On draft day those Yahoo projections and rankings are the default list that will appear in front of most drafters. That means players Yahoo is high on will appear higher in the order for most managers and are more likely to be drafted whether because of auto-drafters or managers seeing the names sitting on their draft boards. The opposite is true for players that Yahoo is low on. The "out of sight, out of mind" phenomenon where players are forgotten or can be had later in the draft at a bargain.

This week on Frozen Tool Forensics: Drafting Yahoo

For the purposes of this article, we are going to focus mainly on points. Given the wide range of league set-ups we may try and hit a few other stats on later articles, but to start, points. As far as process goes, Yahoo doesn't make it an easy export, but with some copy, paste, and column manipulation it is possible (if tedious) to get the data in a usable form. Luckily Frozen Tools is much easier and if we export the Big Board report for 2019-20, we will be able to compare point paces from the 2019-20 season to Yahoo's projections for 2020. We are using point pace as both seasons are going to have an abnormal number of games, and even in the best of times players often play radically different numbers of games. If you play in a season-long roto format you will have to adjust some of this conversation by the number of games you expect a player to get.

One final piece of context here. Points per game can be a little hard to visualize when we are used to full-season point totals. The following table puts the point-per-game difference into total points over a full season.

So a player who is putting up 0.15 more points per game would be improving about 12 points over the course of a season, so let's say jumping from a 65- to a 77-point pace.

But enough process, let's dig in. First up we have players who Yahoo is projecting to do worse in 2020.

So if you are looking for an answer to whether or not Bryan Rust can keep his points, Yahoo is betting on no. Rust blew his 40 and 45 point paces out of the water in 2019-20 and put up 1.02 points per game, or an 83 point pace. Yahoo is betting on a bit of regression and has him at 0.7 points per game, or a 57-point pace. That is a drop of 26 points, and the largest drop Yahoo is predicting. The important question here is, is Yahoo right?

Rust's career season came on the back of a huge promotion in ice time. He went from just under 16 minutes in 2018-19 to just under 20 in 2019-20. More than two and a half minutes of that total time was on the power-play. Even-strength linemates like Jake Guentzel, Evgeni Malkin, and Sidney Crosby are great too, but he also saw time with them in 2018-19. The big difference was the power-play and his 17 power-play points. He ended the 2019-20 season on the second unit, which doesn't bode well for the 2020 season, but now that Patric Hornqvist is in Florida, a top on the spot unit is definitely open.

As far as underlying numbers, his personal and 5-on-5 shooting percentage and IPP are too high. Regression makes sense here, though I think Yahoo is regressing a little too hard. I think low 60's is a pretty reasonable expectation for Rust, which puts Yahoo maybe 5-7 points low. It could make him reasonable value based on Yahoo's rankings.

Yahoo is fading Dougie Hamilton hard in 2020. They are taking his 70-point pace from 19-20 and dropping all the way down to a 50-point pace. In fairness to Yahoo that is pretty consistent with his historical point totals, but his 2019-20 season looks pretty sustainable. He gained 30 seconds of power-play time on ice over his personal best and saw a personal best total time on ice by at least two minutes. His personal shooting percentage was a touch high so knock a couple of goals off, but I still like Hamilton for low to mid 60's, which is a definite improvement over Yahoo's projection. Translation: if your league is drafting based on Yahoo's rankings, Hamilton is great value as he looks poised to surpass Yahoo's projection. 

And now on to players, Yahoo is pegging for a rebound.

Yahoo is really into the bounce-backs in 2020. All of the players in the top ten here are projected for increases of more than 0.2 points per game, which was only true for half of the first list. Quite a lot of the players who we saw struggle in 2019-20, Yahoo is projecting for a return to form. Most of the Nashville and Dallas top lines are included here (plus Matt Duchene just off the list), and notable disappointments Phil Kessel and Johnny Gaudreau are also here.

Dallas first. Yahoo is liking Alex Radulov and Tyler Seguin for a return to form. Seguin is already going to be faded a bit on draft day as he is injured for the start of the shortened season. I do like Seguin to bounce back, his personal and 5-on-5 shooting percentages were low, but I do have concerns about his loss of ice time. He lost 25 seconds of power-play time and more than a minute and a half of total time. Seguin should be better than 2019-20, though I would bet more on a low 70's pace rather than the predicted 77-point pace if that time on ice holds. Radulov, on the other hand, I am steering clear of. The 34-year-old winger lost three minutes of total ice time. That is a huge deal. Without that time there is no way he returns to a 67-point pace. My guess is closer to mid-50's. Let someone else draft Radulov as a near 70-point winger. The upside is possibly there if everything returns to past seasons, but it is a pretty big risk at this point.

Ditto all of that for Phil Kessel. He is 33 and while he hasn't lost three minutes of time, he has lost Sidney Crosby and crew in his move to the desert and it has hurt him. There is a bit of room for his underlying numbers to bounce back a bit, but I think we are talking low 50's not 65.

As far as Nashville goes, I am not generally that optimistic. Viktor Arvidsson, Ryan Johansen, and Matt Duchene are all projected for bounce-backs here but again much of the issue seems to be loss of time on ice, to the tune of two to three minutes. Matt Duchene is the only one whose underlying numbers look reasonable for a rebound (though to maybe a 60-point pace rather than a high 60's). His personal and 5-on-5 shooting percentages look low enough to be outliers. Johansen and Arvidsson, on the other hand, don't seem to have a lot to offer. They didn't play like they deserved a lot more in 2019-20 and the hope with them would be that they revert back to previous playing styles.

That is all for this week. Thanks for reading. Stay safe out there.

Want more tool talk? Check out these recent Frozen Tool Forensics Posts.

Frozen Tools Forensics: Five-on-Five Shooting Percentage

Frozen Tools Forensics: Even-strength IPP and Regression

Frozen Tool Forensics: Home Sweet Home Part II