H2H Matchup Philosophy

[fantasyhockeygeek]

I've got a question that doesn't really fit into any of the forum topics specifically, so it's going here because it's closest to a 1-year idea.

I generally have a few spots on an H2H roster to rotate through hot guys on the wire in order to gain a schedule/GP advantage. In a tight matchup, those decisions make or break your success.

So, from a concept standpoint, how do people approach it? Do you load up players on the team where your opponent holds the goalie (ie if he has a good game and pulls ahead you mitigate the gap) or do you load up on players on the opposing teams (leverage the gap should the game go poorly for his guy)? I find myself stuck on this all the time, so hopefully there are some good thoughts out there.

[hockeyrobot]

So, from a concept standpoint, how do people approach it? Do you load up players on the team where your opponent holds the goalie (ie if he has a good game and pulls ahead you mitigate the gap) or do you load up on players on the opposing teams (leverage the gap should the game go poorly for his guy)? I find myself stuck on this all the time, so hopefully there are some good thoughts out there.

I do not really focus at all on who my opponent owns. Most my leagues have max moves per week and per year so that just takes up too many moves for me. Plus, in a week where teams are playing 2-4 games, the benefit could be minimal to absolutely zero.

I definitely try to leave at least one spot open though as you said. I think that's a great strategy.

[fantasyhockeygeek]

I do not really focus at all on who my opponent owns. Most my leagues have max moves per week and per year so that just takes up too many moves for me. Plus, in a week where teams are playing 2-4 games, the benefit could be minimal to absolutely zero.

I definitely try to leave at least one spot open though as you said. I think that's a great strategy.

Right, so if I'm looking at my roster and one of my churn players for the week will stay on the bench due to schedule (ie I'm starting better players), and there are two players with about the same expected production available on the waiver wire, who I can slot in for 2GP on my roster, who do you pick up?
- Player A, who is on the same team as your opponent's starting goaltender
- Player B, who is against your opponent's starting goaltender

If your opponent's goalie has a bad game, Player B will provide you more edge in your matchup. If your opponent's goalie has a good game, Player A will prevent your opponent from pulling away too far. What process do you guys go through to choose?

[Kofax]

Right, so if I'm looking at my roster and one of my churn players for the week will stay on the bench due to schedule (ie I'm starting better players), and there are two players with about the same expected production available on the waiver wire, who I can slot in for 2GP on my roster, who do you pick up?
- Player A, who is on the same team as your opponent's starting goaltender
- Player B, who is against your opponent's starting goaltender

If your opponent's goalie has a bad game, Player B will provide you more edge in your matchup. If your opponent's goalie has a good game, Player A will prevent your opponent from pulling away too far. What process do you guys go through to choose?

I don't usually follow this line of thinking because even if scenario "good game" or scenario "bad game" plays out for the goalie, there's no guarantee that Player A or Player B will ultimately contribute in that game as well.

I think what you need to do is evaluate all players individually. Go based on how you think that player and his team will perform and pick the player you believe to have the highest probability of contributing.

Now the one situation where I might consider it, is if you are down late in a matchup and need to make up a lot of ground, then I can see the logic in going after the Hail Marry situation of a player against his goalie and hope you hit. Otherwise I think I'm just trying to determine who the best option is regardless of my opponent that week.

[GSD]

I look to the number of games. How many games I can utilize them. What are the chances they may continue on their hot streak. If my opponent has a hot goalie, I would tend to shy away from picking up a player going against a hot goalie, or visa versa. In short I look at the skaters match up not my opponents goalie.

We we are allowed 3 moves a week. I usually have an early week pickup ( mon thru weds). Hot player at least 2 games. Then a late week pickup (same thought process, especially if they play Sunday)Third pickup is for upgrade going forward.

[ericdaoust]

I always worry about myself, not so much the other guy.

[fantasyhockeygeek]

This is an incredibly competitive league so there are almost no matchups that aren't close.

Thanks for the thoughts guys, it's helpful to understand how others think about these things. While I generally agree that every short-term pickup is a crap shoot, I try run through a scenario analysis to risk-manage all my matchups so I systemically increase the worst-, expected-, and best-case scenarios.

[Eskimo Brother]

I don't usually follow this line of thinking because even if scenario "good game" or scenario "bad game" plays out for the goalie, there's no guarantee that Player A or Player B will ultimately contribute in that game as well.

I think what you need to do is evaluate all players individually. Go based on how you think that player and his team will perform and pick the player you believe to have the highest probability of contributing.

Now the one situation where I might consider it, is if you are down late in a matchup and need to make up a lot of ground, then I can see the logic in going after the Hail Marry situation of a player against his goalie and hope you hit. Otherwise I think I'm just trying to determine who the best option is regardless of my opponent that week.

I'm with Kofax. My matchup strategies are based off of trying to win more categories than the other guy. So going into the season, I will try to shape my team around winning certain ones every single week. Then you hopefully have a good enough team to have some 50/50 variance on winning the other categories, and pull out ahead about 80% of the time. If you notice an opponent has done that with a category or two that you don't focus on, your add/drops for the week can punt that category, stacking up on the other 75% of the stats, and gaining an advantage there.

Since head to head is so random based on small sample sizes, trying to gain an edge based on independent variables (the skater/goalie) in the same tiny sample size (one head to head game), doesn't seem the best way to do it. The best way to optimize your chances in small samples sizes, is to give yourself the best odds (better players) and to play with a bigger sample size (having players with more games).

The other point I will make, is that in leagues with goalie categories along the lines of W, GAA, Sv%, SO, I take the following approach: ignore SOs (too random), punt wins (still decently random, and in the end not worth it), win GAA/Sv% 50+ percent of the time by having one good goalie that gets you 2-3 starts, and then stream in another start as needed. This allows you to basically look at the goalie stats any given week as a 50-50 wash, while having an extra skater or two to give yourself better odds in the skater categories (and there are always more skater categories).

However, sticking with the goalies, as soon as stats such as saves, games started, etc. get involved. I like to own a minimum of three starters. That way you can win Ws, Svs, etc most weeks, split the GAA/Sv%, and you only need to tie the skater categories. If done right, it all works beautifully

[lucafen4]

I usually look for the 2-3 players (If 3) 2F 1D that are the hottest at their position, hold multi-positional status and if you count +/- won't wreck that cat. Later in the week, I'll hunt for players based on which cats are closest to push them into wins.