#1 By: Adam Eraky, November 27th, 2018 19:40
Just trying to figure out my squad for playoff matchups, and I noticed that the Buy Low machine has SF RBs (Breida) as a sell for weeks 14 & 15, with an awful playoff slate of matchups, while the fantasy streaming app has him as a top 10 option during weeks 14 & 15. Additionally, Nick Chubb is a top 15 option in the streaming app, but listed as a sell in buy low.
I realize that the apps are calculated differently, so it's not necessarily a conflict that their outputs don't line up. But I'm wondering which one you'd take into account moving forward setting lineups. It seems like the answer is the fantasy streaming app, as it's calculated by utilizing the GLSP projections, but I'm curious what others takes are on this.
Thanks!
#2 By: William Russell, November 28th, 2018 08:03
Good point. I was thinking the same as I tried to answer your start-sit question in another thread earlier.
Adding @FF_Contrarian @dcaban & @CKl who can probably shed more light on the various Rotoviz apps and the weekly Stat Explorer.
#3 By: How, November 28th, 2018 10:02
Anybody else think the GLSP/streaming apps generating funky projections? How could Doug Martin project as the 6th best RB this week, between McCaffrey and Gurley?
I'm trying to decide whether to start Riddick, Coleman, or to grab Doug Martin, Yeldon or Hines to spot start this week. I am trying to fill Melvin Gordon's considerably large shoes while he's hurt. Any thoughts, given that I'm worried something's up with the apps.
#4 By: Shawn Siegele , November 28th, 2018 13:24
Hey @aweraky That's a good question. As you point out, the Streaming App uses a combination of defensive metrics and offensive projection (including the GLSP projections from @dcaban's Weekly Stat Explorer). So it gives a different answer than the Buy Low Machine, which is giving you purely defensive numbers and helping you compare that to other schedules (and helping you get a sense for a player's upcoming schedule compared to his previous schedule).
In terms of how to prioritize them, it depends a lot on the exact objectives. The Buy Low is going to give you a great sense of how players of similar ability might perform and give a sense of how much a player's results have been juiced or limited by his schedule to this point. The Streaming App is going to provide a better look at how you might compare players with different expectations. How much might schedule close the gap between a mid-RB1 versus a low-RB2 or an RB2 versus a low-flex.
That helps to explain why the Streaming App still wants you to buy/play Breida and Chubb despite less than ideal playoff schedules. In this case it's basically saying, yes, schedule isn't great, but don't let that be the only thing you consider.
@will Thanks for the note, and thanks for your kind words on Thanksgiving. (Hopefully you had a chance to listen to the OT pod last week where I mentioned that this community is one of the main things I'm thankful for this year, and you've been a huge part of that.)
@HowFantasy I passed this along to @jim_kloet to make sure the apps version of these projections are working properly. (There was a mild glitch last week that Jim caught quickly and re-ran the calcs.)
#5 By: How, November 28th, 2018 13:35
Thanks! Yeah, for example, it seems to be showing 2012/2011/etc. for the GLSP projections. Strange. For example, the GLSP app shows the below for McCaffrey (note that the streaming app suggests McCaffrey's projections are much stronger):
#6 By: Shawn Siegele , November 29th, 2018 11:35
@HowFantasy Jim has the Sim Scores back up. Let me know if you have further problems. Thanks for bringing it to my attention.
#7 By: How, November 29th, 2018 12:46
Thanks for updating! Still seeing suspiciously odd projections for some players (Theo Riddick projected to dominate, while McCaffrey's projections still look terrible—the quartile score is below negative two!). I'll try other tools for now...
#8 By: Shawn Siegele , November 29th, 2018 12:58
Thanks, @HowFantasy. Jim is doing a ton of work on these and continuing to refine how the matches work and appreciates the feedback.
#9 By: How, November 29th, 2018 16:46
I really appreciate all the hard work. Just to be clear, I'm not trying to dig here. I just want to help by giving feedback when I see things that seem off. I've had such great experiences with the sim scores, with them being so much more predictive and helpful than almost any other tool I've used on any site, that I immediately assume there must be a glitch when the numbers don't seem to add up. The tool is great and very helpful most weeks.
#10 By: thickfreakness, December 5th, 2018 17:46
Is Trubisky missing from the Fantasy Steaming App because of his recent injury?
Powered by Discourse, best viewed with JavaScript enabled