Preseason Blog Series: A new Floored Baby and can you find Breakouts through draft value?

Michael welcomed baby Ruth Ellis into the world last week. Big Brother George loves his little sister…but only for about 5 seconds at a time.

--------

We’re coming down the home stretch of the fantasy baseball offseason. Our draft is set for Sunday, March 28^th, making keepers due on Thursday, March 25^th. We’ve already had 2 big keeper trades and there are still a number of moves available to be made. Keith is the manager that remains most in need of a keeper upgrade and 4 or 5 teams have a spare keeper that becomes worthless in a week…so the time is ripe for Keith to get a pretty good keeper for pennies on the dollar.

--------

The final draft strategy article of the offseason is about how to find breakout players. This blog has talked at length about the value of breakout players: they win you the league. Whether you find them in the draft or on the free agent wire, if you can find players who vastly outperform the price you paid to get them, you will win (hello, keepers that you ‘draft’ in the 11^th round that are 1^st and 2^nd round players).

I had the thought experiment: is there a correlation between draft day value and whether or not a player will break out. Remember, the way I am identifying draft day value is a player who is projected to put up stats better than their average draft position is valuing them at.

Think about it, if DJ Lemahieu is projected to hit 20 home runs with 7 stolen bases and a .290 average…but Jeff McNieil is projected to do the same but at a draft price 60 picks lower, that makes Jeff McNeil a value. This is the whole idea of using projections to find draft day values.

But as we’ve noted in the blog, projections and draft values only take you so far, you need to find breakout players and avoid bust players to win the league. So how do we do that? One thought I had was to try to find a correlation between these draft day values and how often they breakout, so I looked at the projection and end of season data from 2018 to 2020.

Criteria:

·         A draft day value: someone projected 5 or more ranking spots from than their Average Draft Position (ADP)

·         A draft day flat value: someone projected between 0 and -5 ranking spots from than their ADP

·         A draft day not value: someone projected worse than -5 rankings spots from their ADP

·         Breakouts: as discussed in previous blogs, someone who performs 10% better than their projected stat lines at end of season

·         Busts: someone who performs 10% worse than their projected stat lines

·         Flat: someone who performs within 10% of their projected stat lines

Running the data for the various projection systems (Steamer, ATC, THE BAT, DepthCharts, and Yahoo), pulling the average draft positions from each year from either Yahoo or the National Fantasy Baseball Championships website, and comparing the numbers to each year’s end of season data, here is how everything played out:

	Draft Day Value			Draft Day Flat			Draft Day Not Value
Hitting	%breakout	%flat	%bust	%breakout	%flat	%bust	%breakout	%flat	%bust
2018	43%	34%	36%	18%	20%	16%	39%	46%	48%
2019	46%	40%	46%	13%	15%	9%	41%	45%	45%
2020	28%	28%	39%	2%	10%	20%	69%	62%	41%
AVG	39%	34%	40%	11%	15%	15%	50%	51%	45%
Pitching	%breakout	%flat	%bust	%breakout	%flat	%bust	%breakout	%flat	%bust
2018	7%	8%	10%	6%	5%	6%	87%	86%	84%
2019	6%	16%	19%	28%	29%	3%	58%	71%	78%
2020	24%	36%	35%	19%	7%	7%	57%	57%	59%
AVG	12%	20%	21%	17%	14%	5%	67%	72%	74%

Unfortunately, there isn’t too much of a correlation to be seen here. You notice that there are far higher percentages under the Draft Day Not Value columns, but the breakout/flat/bust underneath them are very well spread out. There are higher percentages under Draft Day Not Value because in a player pool of around 1000 players, but only 300 or so players that get drafted and have an Average Draft Position, some players ranked in the 500s may be projected better than those that get drafted in the 200s…leading to the player getting drafted in the 200s being a bad draft day value…but the inverse is very rarely true: the best projected players rarely don’t go undrafted.

What we were looking for here is for there to be higher percentages under the % breakout or bust columns of the draft day values or not values…but that isn’t the case. There does not appear to be any correlation between whether or not a player is a value on draft day and whether or not they will break out. Drafting "Draft Day Value" players, i.e. the Jeff McNeil's of the draft room will still give you a leg up on the league after draft day....they just won't get you a title on their own.

In future blogs we may try to find a better statistic that has a correlation to find breakout players, but that may have to wait until next offseason. Until then, keep shooting your shots to find the breakout players...but try to make them value picks along the way.