Alright guys, you know the drill. Let's talk about the golf course briefly and then some weather, and finally the CORE Plays below. 

Course Preview

Hopefully you checked out this week's Livestream/Podcast highlighting the course. This is vastly diferent to what we saw last week and even the 3 weeks prior. TPC River Highlands is much shorter with many more mid irons and wedge approach shots. Driving isn't too dissimilar but I will highlight the course and what I focused on in PlayerIQ. 

Drives 250-270 (15%)
Drives 270-290 (35%)
Drives 290-310 (50%)

You may ask yourself, why would I use 50% for 290-310 when it was only 42% of tee shots. Well 2 of the 270-290 were nearly right at 288-289, so figured I would bump up 290-310 and lean into that a little more. Historically, nearly ever course is seeing longer drives so if I wanted to place a bet on Driving buckets, it would be that we see a few more in the 290-310 bucket.

Okay moving on. Approach is pretty simple. I want to reiterate why ApproachIQ is so valuable. When you look at an aggregate number like SG:APP (or even SP:APP to be fair), this incorporates all shots from all locations and all distances. It simply doesn't have carryover between ranges. You can view any players profile and see that not all ranges are created equal. Some players excel with the long irons while others excel with wedges. Even if a player like Scottie is phenomenal with all irons, you will still see he excels "more" in certain buckets than others. Distance aside, location also matters. If approach isn't noisy enough, shot locations pose even more risk. 

Especially coming off a week like Oakmont where guys literally lost full shots on single holes because of the rough, it simply is unfair to weight these guys equally to players who hit more fairways. And don't worry, we aren't chasing our tail, poor players would be identified in the DrivingIQ research. It is simply unfair to use all shots from a betting or DFS persective, this same principle holds true for why Spectrum data exists in the first place. We aren't trying to award season long performances for what players have done, we are trying to be as predictive as possible and "cleaning" the data in order to have accurate useful data is incredibly important.

Not to go on a tangent, but even Proximity, as simple as it sounds: "How close does a player hit it on average from x distance", there are too many outliers, even if focused on proximity from the fairway. I played golf today and was tracking some "real-world" metrics, and proximity from 100-125 yards was skewed because of one poor approach shot to 35'. It didn't matter that the other 3 approach shots fro that distance were inside 15', now my average has gone from 10' to 16' just because of 1 shot. 600% increase.  All of this is to say, understanding the data and the pros and cons of different splices is very important.

Okay back on topic, the approach shot distributions are much different than what players have seen recently. You can see this in the ApproachIQ data for