I've taken some detours lately, but today I'll be returning to my series examining each of the primary daily fantasy sites, their roster structures, scoring systems, and the types of players that are most useful for each particular site. If you haven't heard yet, the subject of a previous entry in the series, Daily Joust, is no more.  As such, the metrics have all been recalculated using the new eight-site average, if you're wondering why they look a little different. Previous Profiles: Draft Day | Fan Throwdown DraftKings Profile Uniqueness: 1st (Hitting), 6th (Pitching) Scoring System Quirkiness: 5th (Hitting), 2nd (Pitching) Roster Quirkiness: 3rd (Tie) Pitching Premium: 4th DraftKings is one of the big daddies in the daily fantasy landscape, so I thought it was about time we see how they differ from the competition. Uniqueness Uniqueness is a measure of how differently players score on a particular site relative to the average daily site. Uniqueness serves as a good estimate for how much your strategy needs to change when playing on a given site. (Note: Pitching Uniqueness only includes starting pitchers despite some sites utilizing relievers.)

Site	Hitting Uniqueness
DraftKings	3.7%
	3.6%
	2.9%
	2.7%
	2.6%
Fan Throwdown	2.5%
	2.2%
DraftDay	1.7%

DraftKings is one of those sites like DraftDay where their Hitting and Pitching Uniqueness are almost polar opposites.  Except for DraftKings, it's reversed.  With the former most unique site, Daily Joust, no longer in the mix, DraftKings takes down the title of most unique site for hitting.  That is to say that hitters score far more differently on DraftKings than they do on any other site.  In terms of pitching, though, DraftKings pretty much toes the line established by the other sites.   Scoring System Quirkiness (Hitting) Scoring System Quirkiness is a measure of how much a given site's scoring system differs relative to the average daily site. Since sites award points on different scales (i.e. DraftDay gives 20 points for a HR; FanDuel gives 4 points for a HR), category values are listed in relative terms (all relative to the number of points awarded for a home run. So if a single is worth 1 point and a home run is worth 4 points, the value of a single is said to be 25 percent.)

Category	DraftKings	Average
1B	30%	26%
2B	50%	50%
3B	80%	76%
HR	100%	100%
RBI	20%	31%
R	20%	27%
BB	20%	22%
SB	50%	50%
CS	-20%	-12%
HBP	20%	20%
K	Not Used	-5%
GIDP	Not Used	-4%
SAC	Not Used	2%
Out	Not Used	-3%

DraftKings is the only site to eschew the 25/50/75/100 relative scoring format for 1B/2B/3B/HR.  Instead, DraftKings places heavier emphasis on singles and triples and less relative emphasis on the long-ball. The site also places less emphasis on RBIs and Runs than other sites, particularly RBIs.  Finally, it's one of four sites to use Caught Stealing as a category, but it uses the softest penalty of the group (the other three sites all use -25% scoring). While DraftKings is only moderately quirky when it comes to its hitter scoring system, those quirks pack a lot of pow in their punch.  The emphasis is clearly on low-power hitters since the importance of HR is reduced relative to other hit types and since the importance of RBIs is reduced relative to runs.  This would indicate that speedy slap hitters that wrack up a lot of singles and, when they go for extra bases, triples are more valuable on DraftKings.  The use of Caught Stealing cuts against their value a bit, so you're going to want to focus on these types of hitters that either don't steal a lot or are successful at a high clip.  Just be careful because Stolen Base Success Rate (SB%) is a highly variable stat.  You need years of it before you can be sure that a player is actually good/bad and not just lucky/unlucky.  Also be careful not to chase triples too much since they are a relatively rare occurrence and their relative value isn't that much higher. And since the value of both RBIs and R are diminished, it would also be wise to target guys who have good skills but who don't have a good lineup spot. So the combination of high-average, low-power, and less-than-optimal lineup spot (the last one being the most important) leads to guys like David Freese, David Murphy, David Lough, Chris Johnson, and James Loney. Scoring System Quirkiness (Pitching) Scoring System Quirkiness is a measure of how much a given site's scoring system differs relative to the average daily site. Since sites award points on different scales (i.e. StarStreet gives 15 points for a win while DraftStreet gives just 1.5 points for a win), category values are listed in relative terms (all relative to the number of points awarded for a win).

Category	DraftKings	Average
W	100%	100%
Outs	19%	13%
K	50%	30%
ER	-50%	-31%
H	-15%	-8%
BB	-15%	-8%
HBP	-15%	-7%
SV	Not Used	38%
BS	Not Used	-10%
CG	63%	26%
SH	63%	14%
L	Not Used	-15%
No No	125%	22%
PG	Not Used	6%

Despite a low Pitching Uniqueness score, DraftDay is near the top in terms of Quirkiness.  This has a lot to do with their use of bonus stats like Complete Games, Shutouts, and No-Hitters.  As we saw with Draft Day, though, these stats don't affect our decision-making much since they're so difficult to predict. Instead, we want to focus on the way DraftKings strays far from the norm when it comes to most of the normal categories.  We see that they reward a lot more for strikeouts and generic outs (aka innings pitched) but that they also punish a lot more for hits, walks, and HBP.  This has the general effect of evening things out, which results in a low Uniqueness score.  You could target low-walk pitchers who pitch deep into games, but you're not going to gain much of an advantage since they're going to lose relative value by not striking out a lot of batters and allowing more hits.  What we really need are high-strikeout, low-walk pitchers that go deep into games.  These pitchers are otherwise known as aces. Yup, it's kind of boring, but you want to place more emphasis on elite pitching when you're playing DraftKings while ignoring lesser pitchers.  Matchups are of course still important and can make a lesser pitcher playable, but relative to other sites, sound DraftKings strategy involves targeting the best pitchers each day. Roster Quirkiness Roster Quirkiness is a measure of how the roster structure of a given site differs from the average daily site.

Position	DraftKings	Average
C	1	1
1B	0	0.4
1B/DH	1	0.5
3B	1	0.9
1B/3B	0	0.3
2B	1	0.9
SS	1	0.9
2B/SS	0	0.3
OF	3	3.0
U	0	0.8
SP	2	1.4
P	0	0.6
Total	11	10.8

We see that DraftKings is tied with the subject of one of our previous installments: Fan Throwdown.  The two, along with one other site, have the exact same roster structure.  With the exception of not using a Utility player, DraftKings is pretty much average everywhere else.  Without the Utility spot they wrap DHs into the 1B spot but then use the league average setup at every other position.  Pretty boring, but it does give a slight boost to DHs since they only have first basemen to compete with to make your roster and not the entire pool of available hitters.       Pitching Premium Pitching Premium is a measure of how valuable each pitcher spot is relative to each hitter spot based on each site’s scoring system. So on Fan Throwdown, for example, the average pitcher is 47 percent more valuable than the average hitter.

Site	Pitching Premium
	76%
	75%
	71%
DraftKings	62%
DraftDay	49%
Fan Throwdown	47%
	34%
	25%

DraftKings is pretty boring here as well, sitting close to the middle in terms of Pitching Premium, right on the bottom edge of the top group of sites.  The site places a slight premium on pitching relative to other sites (average is 55%), which is another log on the "draft elite pitching" fire since it's slightly more important to get a good pitching performance than it is on other sites.       Methodology for Creating Each Stat If you’re like me and want to know what goes into the sausage, here is how I arrived at each stat that I created. If you don’t care, then you’ve reached the end of the article. You are now free to leave. Uniqueness is calculated by first looking at how every player scores in each of the eight systems. They are then re-calculated on an index scale, comparing each player to the site’s average player so that all sites are using the same scale. Then a new average is created for each player of his score on each of the eight sites. We examine how far each site’s score is from the eight-site average as an absolute value. When we average these out for all players for all sites, we get an estimate of each site’s “uniqueness.” I used full-season data for all hitters with at least 300 PA, all pitchers with at least 10 Games Started in 2012. Scoring System Quirkiness is calculated by first putting all stats on the same scale (relative to Home Runs for hitters and relative to Wins for pitchers). I find the eight-site average value of each stat, compare how each stat’s scoring varies from that average, then average out all the categories for each site (with each category weighted the same). Roster Quirkiness is calculated by finding the eight-site average of how many players are required at each roster position, then comparing how each site’s roster structure varies from that average, and average out all the roster spots for each site. Pitching Premium is calculated by first scoring out the average daily line for all hitters with at least 3 PA in game (a proxy for starting the game) and all starting pitchers for each of the eight sites. I then compare how much more valuable pitchers are than hitters using data from a recent six-year stretch of games.