How you can use your fantasy baseball roto league history to plan out your draft. If the sabermetrics revolution has taught us anything, it’s that as fans, we really don’t know what the heck we’re talking about. Various forms of bias creep into our minds that really have no statistical significance or can be proven totally wrong with a few numbers. Since playing fantasy baseball is a pretend fantasy of actually managing a baseball team, we’re allowed to totally screw things up. No one is going to fire you unless you’re one of those people who auto-drafts a team and then never looks at it during the season, don’t be that person. You can choose a player because he’s on your favorite team. You can say to yourself, “I think this guy’s totally going to break out this year”, without doing any research or having the data on a progression of skills to back you up. You can take who the average draft position (ADP) says you should take. But, none of those things will set you up to win. This year, I’m approaching my drafts like a robot who knows nothing of fandom or human connection, only numbers and stats. I’m not going to make any gut calls, any fan picks, or any huge gambles. I’m going to look at what the numbers say based on projections built on statistical models and I’m going to build a winning team. Want to join me, fellow stat-head robots? Here’s the easiest way to set yourself up to make the right picks going into your 2021 draft in 3 easy steps. Step 1: Don’t pay (much) attention to ADP Some ADP measurements can be good, like that on the National Fantasy Baseball Championship site, mostly because these are high dollar fantasy leagues and it cuts out bias a little more when the stakes are so high. Beyond that, ADP is mostly just telling us where other people did their drafting on average. Included in that ADP measurement are people’s biases, fan hoods, and weird, unexplainable expectations. People draft Player A because they really, really want him to be good. Even if Player A has never shown any of the necessary skills to be good at the major league level. Don’t let other people’s biases, hopes, and dreams affect the way you build your team. Step 2: Use projections There are many great projection systems in use today with some of the most popular being FanGraphs Steamer or Depth Charts, PECOTA, Pod Projections, and THE BAT-X. Each has its own variation of projection style, but any of these systems can be used for gaining an edge. This is not to say that there’s anything wrong with ESPN’s projection systems, but the projections listed above are typically more analytically driven, released way before ESPN’s and if everyone else is using conventional and you’re using synthetic...well...you have an edge, however small. I would recommend researching these systems and finding the one you like best or is most easily available to you. But if you’re all like, “Naw man, I don’t want to do research, just tell me what’s up.”, then I would say use Depth Charts from FanGraphs. Step 3: Calculate your league averages and find your players Now we get to the real meaty part, finding out how many stats you need to realistically contend for a top spot at the end of the year. Each season you’ve played in your league, there has been a winner. If you’re joining a league for the first time, well, there were still winner’s in the past. If it’s a totally brand new league, there certainly were leagues like it in the past and you can find them by doing some searching around whatever platform you're using. First, start by finding all the winners and their year-end stats. For example, my ESPN roto 5x5 (.obp instead of .avg) has been going for 5 years. I excluded 2020 from my averages because of the shortened season to get the following averages: R: 1030 HR: 316 RBI: 971 SB: 170 OBP: .360* * Use the calculation for OBP to help you figure out how many hits/walks/ABs you'll need, math can be fun! What you can do from here is figure out, depending on your roster size, how many of these counting stats you need for each of your rostered players. In my case, my offensive roster size is 11 players, so I need something around 94 runs per player: 1030 (total runs from above) DIVIDED BY------------------------------------------------------------------------- 11 (number of offensive players on my team) Obviously, that would be hard to do. You won’t be able to snag 11 players who will cover all of these stats. Maybe you can go after a top 10 run-scorer to balance out a player who might be under that 94 target. However, your goal is to always be around the projected total of 1030 runs from your starters when drafting.
One of the best things this draft style does is it eliminates the one stat players from your draft pool. Yes, you can pick up a shortstop who is expected to steal 25 bases, but that’s all you’ll get. Instead, look for a bunch of players who can steal around 10 bases. Some can project for 12 and some can project for 8, but you’re adding value with each player and trying to get as close to those necessary totals as possible. Remember, you’re using your projections to figure out what you can expect from a player. So go find the players who are well balanced and you’ll be surprised to see who pops up. So how will this help you win? The biggest reason you should use this system is to eliminate bias. As sports fans, we often throw out crazy, uninformed numbers. In the past, you’ve just reached for players because you had them the year before and they did pretty good, but you might have missed that article about the very visible skills decline with age that player has been showcasing. You might see someone who will steal higher than average bases but is empty in almost all the other stats. Don’t fall for it, let someone else take him. You are better prepared than most. In a way, this can be tough. We’re all fans of the game and we all have players that we lean towards because we just like them. But, if your projections are telling you what you need to know, being a fanboy won’t get you to that sweet, sweet victory stage, will it? No, it won’t. Instead, be like me, be a robot-machine-computer and take the stats and in turn, the trophy!
0 Comments
Leave a Reply. |
AuthorThis blog is dedicated to baseball analytics and general baseball discussion. Archives
January 2021
Categories |