Good to Be Lucky and Lucky to Be Good

Using Expected Goals to find lucky and unlucky players to buy in to in fantasy.


If you are a fan of fantasy baseball, like I am, arguably more than hockey (don’t tell the others) you probably have heard of expected statistics before. In baseball, Expected Batting Average (xBA), Expected Weighted On-Base Average (xWOBA) are some popular ones that each hold their own value. Hockey, although lacking behind baseball in terms of advanced analytics, do have their own set of expected stats. The most important for fantasy hockey would likely be Expected Goals For (xGF). There are several other relevant expected stats to be utilized in fantasy, but I’ll save those for you to explore yourself, or for future articles.


credit: cbc.ca

In simplest of terms, Expected Goals For or xGF is a statistic that calculates the number of goals an individual or team should expect to have given certain variables. It is more complex than a simple Excel formula. It is found using mathematical models far above my knowledge or paygrade. Lucky for us there are plenty of resources to help you find these numbers, several of which FP Mike has mentioned in some YouTube videos. For this article I will turn to one of my favourite sites, MoneyPuck.com.


There are lots of variables and factors that go into MoneyPuck’s Expected Goal Model. A few of the variables are easy to identify and an average fan could assume this event leads to a higher likelihood of scoring a goal such as proximity to the net or shot angle. Other factors are more in-depth and not easily followed during a game such as time since previous game event (hit, faceoff, etc.). All-in-all, the model aims to put the goal expectancy into one number. It is like saying “given the factors of Player X’s shots over the course of a season, this is how many goals we expect Player X to have". Here is a look at a chart comparing Goals vs. Expected Goals. As expected, (no pun intended), as expected goals increase, so do goals.

The of this article is to find potential sleepers, breakouts or comeback players that can be scooped up in drafts or on waivers and help your team in fantasy. In addition to being beneficial to your fantasy league, those with high xGF are also going to be useful in DFS contests.


To start let’s look at the overall leaders in xGF and compare it to how many goals they actually had in the 2019-20 season.

Can you tell why FantasyPuck likes Brady Tkachuk yet? This chart is not too surprising. Four of the five players on this list were also within the top 10 in goals last season.


Now if we want to factor in some luck, you could say Ovechkin had some good luck. A difference of about 14 goals is not completely up to luck though. Ovechkin is still the best shooter in the league and clearly gets himself into good positions to score on a consistent basis. On the flip side, Tkachuk likely faced some bad luck this past season. There is a variety of factors you have to consider when evaluating a player on their xGF, some are listed above from MoneyPuck and some are not considered such as the goaltender they were facing. Either way, expected goals are a good thing. Expected goals equal offensive production for that player which is what I am looking for on my fantasy teams and in DFS.


Let’s look at a couple of players I like based on the difference between their xGF and their actual goals.


Brady Tkachuk:

I know, I know. Don you guys talk about him way too much at FantasyPuck. That’s because he has so much to like about his game. Imagine him on a team like Ovechkin’s Washington Capitals? He would probably be a top-10 pick this season in fantasy as well as one of the highest-owned players in DFS. If you read my article that looked into Rebounds and how we can utilize that information for fantasy, you would have seen Tkachuk shoots a ton and shoots a ton from close to the crease. He led the NHL in high-danger shots (42) and was tied for third in rebound goals (7). When you are shooting that much you are creating goal opportunities. The first variable MoneyPuck lists on their website for what goes into the xGF and shot predictor models is “shot distance from net”. With so many high-danger opportunities and only 21 goals, perhaps Tkachuk could work on his accuracy, but nonetheless, these numbers are too good to pass up. Brady is a guy you need to grab in drafts. The production will continue to increase, especially with the bright future Ottawa has. Given his mediocre fantasy stats such as goals and points, his value will be very high in DFS contests.


Jordan Staal:

The second best Staal brother (behind Jared obviously) may be nearing the end of his career at 32-years old, but proved last year to still be a valuable top-9 forward on a shoot-first Carolina team. 8 goals is not really worth a look in fantasy drafts. He is expected to play 3rd line and will likely see some power play time again as he did last season, but aside from an occasional DFS appearance, he shouldn’t really be relevant for us. Pretty interesting to see in this case how if you were to only look at his xGF, 16 goals would seem like a pretty decent season for a veteran like Staal. He is surrounded by names like Pierre-Luc Dubois and Nico Hischier on the xGF leaderboard. Definitely do not take Staal over either of those two, but I remember at least once or twice stacking Carolina’s third line with someone like Tampa Bay 1 or Boston in a DraftKings contest. Something to consider again this season.


Jack Hughes:

Here is another player with about half as many goals as expected goals. If Hughes finishes with 14 goals in 61 games instead of 7, I don’t people would be using the term “bust” on him just yet. We know of the sophomore slump, so perhaps Hughes decided to reverse his first two years in the league and this season will be productive. On a team like New Jersey, I wouldn’t be betting on a huge season for Hughes as he has next to no support. There is maybe three players I’m considering on the Devils in fantasy drafts, Hughes is one of them. Hughes is being drafted late. I have seen his ADP around 160 and at that point you are usually making bets on players you feel good about. New Jersey is committed to him playing and playing with the best line mates possible (Gusev and Palmieri) or else they would have sent him to the AHL last season. Unless you are in a draft with my sister or other managers who might think he is cute, you can probably get away with waiting on Hughes until the last five rounds.

---

I personally prefer to look at the under performers with the xGF stat. From the few years I have been using this stat to help evaluate players, I have noticed that significantly more goals than expected goals likely means you are just a good shooter, whether they are more accurate or harder shots. There is certainly some luck involved as well, but you need to be good to lucky and you need to be lucky to be good. I still will point out a couple of players that surprised me on the leader board and how I feel their xGF will come into play this season.


The leader board for largest positive gap goes as follows:

All superstars and all players being drafted in the first round. This shouldn’t change your opinion on these players. As far as players who I believe may have been the beneficiary of some good luck, let’s have a look.


Elias Lindholm:

By no means do I think Lindholm is bad or not worth a look in fantasy or DFS. He stepped up in many games last season when the streaky Johnny Gaudreau was under-performing. The questionable piece I notice from Lindholm’s season in 2019-20 was that it was the first season in his professional career (since 2011-12) that he had more goals than assists. I haven’t followed the Flames too closely so I really have no idea if Lindholm has an answer for that obscurity. Perhaps he worked on his shooting in the offseason or over COVID-19 lockdowns? He could also have been benefiting from his line mates Sean Monahan and Johnny Gaudreau. Currently, the lines show him playing with Dillon Dube and Matthew Tkachuk. If I were to guess, I would say Lindholm returns to his 1 goal to 2 assist pace he has shown in the past decade. I don’t mind Lindholm especially if he plays first-line minutes, but if there are more pure snipers around the same area like Kyle Connor, Filip Forsberg or Taylor Hall, take them instead.


Dominik Kubalik:

The sharpshooting rookie Kubalik got off to a slow start before really heating up in December. Kubalik has already exceeded expectations considering he was a 7th round pick in 2013. While he may have surpassed career projections from draft day already, his expectations will likely not be met in 2021. Keep in mind Kubalik played most of his minutes last season with Jonathan Toews. Toews is out of action until an unknown date and the replacements in Chicago are not good. The current lineup has Kubalik playing alongside Lucas WalMart, I mean Wallmark and Pius Suter. Who? Good question, but neither of those guys are going to contribute in the same way Toews does, therefore impacting Kubalik and his production. He is getting drafted around the 130 mark. I’d rather take someone like Lafreniere and bet on him having a breakout year like Kubalik did in 2019-20.


That wraps up this article for me. I want to emphasize having a high or low xGF is not the full story of a player’s fantasy performance as we have seen last season. This is simply a fun and entertaining stat I like to use to find potential diamonds in the rough or players I should look further into. Hopefully I was able to provide some useful advice and either answer some questions or introduce you to expected stats and the benefit it has to us as fans.


-FP Don



Sources: MoneyPuck.com, NaturalStatTrick.com


72 views0 comments

Recent Posts

See All