This article is part of our MLB Daily Games Strategy series.
We are a rule driven society. Every aspect of our lives seems to be governed by written and unwritten rules. As a scientist, I'm supposed to question rules and laws.
What do we really know is true and how do we know it? What if it really was otherwise and we just didn't yet have the data or method to prove it?
Scientists should be trained to think everything they know or believe is potentially wrong. In searching for the truth, you have to rigorously evaluate and discard all other options. This is why you always hear about new scientific data supporting a hypothesis, suggesting a new truth, or occasionally, refuting a previously held conclusion.
We present our results with confidence intervals: how sure are we that what we've discovered is TRUE? Using statistics, 95% is a common threshold used in science. If we are 95% sure that what our experiment reveals is true, we report it to the public and our colleagues for their scrutiny.
Sometimes scientific truths last years, even decades, before technology provides a means to unseat them. An example I give in one of my classes is of birds that migrate extremely long distances, presumably by using the Earth's magnetic field to navigate. A few years ago, it was reported that the magnetic-sensing neurons had been located in the beak of a certain species of migrating birds. However, newer methodologies identified these cells as metal containing, but not magnetic-sensing, returning the mystery of how the birds
We are a rule driven society. Every aspect of our lives seems to be governed by written and unwritten rules. As a scientist, I'm supposed to question rules and laws.
What do we really know is true and how do we know it? What if it really was otherwise and we just didn't yet have the data or method to prove it?
Scientists should be trained to think everything they know or believe is potentially wrong. In searching for the truth, you have to rigorously evaluate and discard all other options. This is why you always hear about new scientific data supporting a hypothesis, suggesting a new truth, or occasionally, refuting a previously held conclusion.
We present our results with confidence intervals: how sure are we that what we've discovered is TRUE? Using statistics, 95% is a common threshold used in science. If we are 95% sure that what our experiment reveals is true, we report it to the public and our colleagues for their scrutiny.
Sometimes scientific truths last years, even decades, before technology provides a means to unseat them. An example I give in one of my classes is of birds that migrate extremely long distances, presumably by using the Earth's magnetic field to navigate. A few years ago, it was reported that the magnetic-sensing neurons had been located in the beak of a certain species of migrating birds. However, newer methodologies identified these cells as metal containing, but not magnetic-sensing, returning the mystery of how the birds detect magnetic fields to the realm of the unknown. The magnetic-sensing beak neurons were a temporary truth.
DFS is new enough that there aren't too many rules yet, written or unwritten. Experienced individual players likely have their own personal set of laws they use to set lineups and enter games. In so much as these get bandied about chat rooms, forums, and on Twitter, some make it to "rule" status. Many of my articles in this and the NBA DFS strategy series have tackled unwritten rules that I've come across. Simply asking the question, "Are we sure this rule is right?" opens up a dialog that can lead to better player and situation evaluations by us DFSers. In some cases, such as the rule to avoid high pace players in "low pace" games, the rule was proven invalid when the actual data was examined.
Why are there so many flawed 'laws' then, of science and sports? For one thing, it's our innate ability to organize information and categorize data that leads to our acceptance of rules before we have sufficient evidence to support them. If we can file away some piece of information as fact, rule, or law, with other similar bits of information, our brains can save time and energy when recalling and using that information. It's simply efficient.
Another factor is peer pressure--if someone we know and respect makes an observation or suggestion, it can quickly take on rule status among his/her friends and followers. If the same observation is made by someone with less expertise, it may be quickly dismissed or dismantled. The "1st game at Coors Field" rule comes to mind.
In DFS one of the few cardinal rules involves the difference between GPP and cash game lineups. The unwritten rule states that your GPP lineup must be diversified in such a way that you'll hit on the top plays that others will miss.
- You must gamble on the players with upside using a Studs and Scrubs approach
- Spend on pitching and scrimp on hitters like Danny Santana or Kevin Kiermaier OR
- Pick a cheap pitcher like Bud Norris or Jarred Cosart to pair with top-notch hitting
- You must be contrarian. Avoid obvious plays that "everyone" will be on.
- The pitcher facing the Padres
- Hitters in Coors Field
- Lonnie Chisenhall for the next three days
Notice these are precisely the plays you'd want to use in your cash game, according to the rule. They are the players that give you the best chance to score fantasy points. Solid, safe, evidence based plays.
Your GPP lineup should therefore be full of low % owned, high risk/high reward players, according to the rule. IF those guys have good nights, you've put yourself in a position to be on top of the pack. Alone.
Unfortunately, I haven't come up with a way of quantifying this in such a way that I can make a graph and show you how successful the lineups are that follow the rule vs don't follow the rule. But I do study lineups in both types of games on multiple sites every night so I'll share my observations.
- MLB DFS is variable enough without your trying extra hard to find it. As a friend of mine put it, each hitter's going to get 4-5 chances at the plate. With that small sample size, anything can happen. Every night some of the safest guys go 0/4. Every night at every position you find the most expensive and cheapest salary players scoring the same fantasy points.
- No one is ever ALL IN on the same player. Occasionally on single pitcher sites you'll see 80-90% ownership if it's a really dark slate. Like once this year maybe. Even last night Steven Strasburg was only ~40% owned.
- The lineups that win GPPs are not full of 1-5% owned players. Last night's DFBC qualifier winner, chripe01, had two low owned players, Adam Eaton and Scott van Slyke (who I also used and can't understand why he wasn't higher than 3% owned). The rest of his lineup consisted of players that were 11-53% owned, Chisenhall being the 11%. This is one example, but a theme I see night after night. Either these winners are not trying to be super contrarian, or everyone else is trying to be contrarian in the exact same ways. But when the lineup includes Freddie Freeman, Jason Kipnis, and Jose Reyes, I don't think it's the latter.
How can you use these observations? Well, I hear people talk about "my GPP lineup and my cash game lineup" all the time. It's a distinction I get for NFL and NBA, which are much more predictable sports. If a high priced starter plays the whole game, he's going to score fantasy points for you 95% of the time, period. That is simply not the case in baseball though.
When I've placed 1st in MLB GPPs, it's with the same lineup that I place 1st in 50/50 and h2h...the best lineup I could construct. So I don't approach my lineup construction with the mindset of the GPP rule at all. When I do use two lineups, it's simply because there are too many players that I want exposure to. I'd then divide my entries pretty evenly between them.
Not using the players you, the experts, and the numbers say have the best chance to score fantasy points in your GPP lineups for any of the reasons outlined above seems silly to me. Being contrarian for the sake of being contrarian in baseball is also silly. Even if it works out once in a while. The one time I make an exception is for multi-entry GPPs. Then you have two interesting choices...diversify your lineups so you cover all your bases so to speak, or multi-entry the same lineup hoping to multiply your winnings if it hits. If you diversify, then you'll be trying to use the logic of the GPP rule and I understand it.
As always, this is how I interpret and play the game. If you track such things, tell me how you guys approach different games and which strategies have taken you to the top of the GPP standings. A "special" GPP lineup or your cash game/best possible lineup? Good luck in all your games this week!
