The DLF Dynasty Trade Analyzer

Administrative announcements, suggestions, opinions, complaints, problems, wish list items. Whatever it is about the administration or operation of the site or forum, you can find or post it here.
jkid2
Starter
Starter
Posts: 547
Joined: Mon Jun 11, 2018 9:04 am

Re: The DLF Dynasty Trade Analyzer

Postby jkid2 » Wed Sep 25, 2019 8:27 am

Can we have adjusted pick values for league sizes other than 12 teams? Obviously picks are worth a bit more in 10 team leagues because a 1st is ranged from 1-10, 2nd 11-20, etc. Is there any way to factor that into the calculator?
10 Team PPR
Roster: 26 Players + 2 TAXI + 2 IR
Start: 1 QB, 2 RB, 2 WR, 1 TE, 2 Flex (RB/WR/TE)

QB: Aaron Rodgers, Matthew Stafford, Tom Brady
RB: Le'Veon Bell, Aaron Jones, Chris Carson, Marlon Mack, James White, Rashaad Penny, Nyheim Hines, Rex Burkhead, Jerick McKinnon (IR), Qadree Ollison, Rodney Anderson (IR)
WR: Michael Thomas, Mike Evans, Odell Beckham, D.J. Moore, Brandin Cooks, Robert Woods, Corey Davis, Andy Isabella, Tre'Quan Smith
TE: O.J. Howard, Noah Fant, David Njoku, Jonnu Smith, Kyle Rudolph
TAXI: Daniel Jones

Draft Picks:

2020: 2X 3rd, 2X 4th
2021: 2nd, 5X 3rd, 7X 4th
2022: 1st, 2X 2nd

User avatar
dlf_kenm
Administrator
Posts: 1274
Joined: Sun May 14, 2006 12:28 am
Contact:

Re: The DLF Dynasty Trade Analyzer

Postby dlf_kenm » Wed Sep 25, 2019 4:01 pm

Thanks for the suggestion jkid - I'll see what we can do.

jonesmaster
Role Player
Role Player
Posts: 330
Joined: Wed Jun 29, 2016 10:02 pm

Re: The DLF Dynasty Trade Analyzer

Postby jonesmaster » Mon Oct 21, 2019 10:24 am

Can you explain the analysis that goes into the package value discount -- like how youre coming up with the reduction in point value?
10-team 0.5 PPR Scoring TE Prem (Full PPR): 1QB, 2WR, 2RB, 2FLX, 1TE, 1D/ST, K

QB: Brady , Rivers, Brissett
WR: Moore, Juju, Evans, Woods, Watkins, Sanders, Hakeem Butler, Humphries, Cobb
RB: Zeke, Gurley, Mixon, Mike Davis, Gio, Ty Johnson, Perkins, Armstead
TE: Kelce, Reed, Jonnu, Moreau
D/ST: Bears, Colts, Panthers

10-team Full PPR Scoring: 1QB, 2WR, 2RB, 1FLX, 1 SFLX, 1TE, 1D/ST, K

QB: Wilson , Mahomes , Baker, Teddy B, Stidham
RB: Coleman , White , Breida, Drake, Gio, Lewis, Davis, Ballage
WR: Evans , MT , Ridley, Harry, Boykin, Robby Anderson, Dede
TE: Kittle , Fant, Rudolph, Reed, Dissly
D/ST: Bills, Colts, Vikings

2020 Picks: 1, 1, 2, 2, 2; 2021 Picks: 1, 2, 2, 3, 3

dlf_mikem
Administrator
Posts: 9
Joined: Wed May 25, 2016 2:37 pm

Re: The DLF Dynasty Trade Analyzer

Postby dlf_mikem » Wed Oct 23, 2019 1:32 pm

Hi jonesmaster. The idea behind the package adjustment is to prevent someone from stockpiling too many lower value assets on one side of a trade versus a smaller number of high value assets on the other side, to tip the scales to the first (higher quantity) side. There should be a cost or penalty to this because roster spots are valuable and, typically, better players carry a premium over a bunch of average meh players.

There were a few different iterations of the package adjustment calculation because it is tricky to figure out given a wide range of possible trade combinations. The Analyzer looks at the number of assets on each side of the deal, the value difference as a % between the two sides, and applies a minimum between a couple of different calcs.

Michael Zingone worked up the math for it, so you can try to hit him up on Twitter (@FFZinger) for more details.

jkid2
Starter
Starter
Posts: 547
Joined: Mon Jun 11, 2018 9:04 am

Re: The DLF Dynasty Trade Analyzer

Postby jkid2 » Thu Oct 24, 2019 8:21 am

We've had some discussion on this and we're struggling with how to quantify the differential in value based on league size alone.

How is the 15th overall pick different in value in a 10 team league than it is in a 12 team league? How do you quantify that difference in a non-abstract, meaningful manner? And ultimately, is there really an actual difference in the value of that pick?

Thoughts on this are welcomed.

Thanks
dlf_kenm wrote:
Wed Sep 25, 2019 4:01 pm
Thanks for the suggestion jkid - I'll see what we can do.
Is there any updated on different pick values for different league sizes? I figure it's a decently simple fix based on average pick value right?
10 Team PPR
Roster: 26 Players + 2 TAXI + 2 IR
Start: 1 QB, 2 RB, 2 WR, 1 TE, 2 Flex (RB/WR/TE)

QB: Aaron Rodgers, Matthew Stafford, Tom Brady
RB: Le'Veon Bell, Aaron Jones, Chris Carson, Marlon Mack, James White, Rashaad Penny, Nyheim Hines, Rex Burkhead, Jerick McKinnon (IR), Qadree Ollison, Rodney Anderson (IR)
WR: Michael Thomas, Mike Evans, Odell Beckham, D.J. Moore, Brandin Cooks, Robert Woods, Corey Davis, Andy Isabella, Tre'Quan Smith
TE: O.J. Howard, Noah Fant, David Njoku, Jonnu Smith, Kyle Rudolph
TAXI: Daniel Jones

Draft Picks:

2020: 2X 3rd, 2X 4th
2021: 2nd, 5X 3rd, 7X 4th
2022: 1st, 2X 2nd

dlf_mikem
Administrator
Posts: 9
Joined: Wed May 25, 2016 2:37 pm

Re: The DLF Dynasty Trade Analyzer

Postby dlf_mikem » Thu Oct 24, 2019 10:05 am

I actually thought we did adjust for that, so, for example, the value of 1.11 (11th) in a 12-team league = value of 2.01 (11th) in a 10-team league, but that is not readily apparent right now because the current season picks are not available. Do you think the values of those should be different? I would have thought the same.

jkid2
Starter
Starter
Posts: 547
Joined: Mon Jun 11, 2018 9:04 am

Re: The DLF Dynasty Trade Analyzer

Postby jkid2 » Fri Oct 25, 2019 8:09 am

dlf_mikem wrote:
Thu Oct 24, 2019 10:05 am
I actually thought we did adjust for that, so, for example, the value of 1.11 (11th) in a 12-team league = value of 2.01 (11th) in a 10-team league, but that is not readily apparent right now because the current season picks are not available. Do you think the values of those should be different? I would have thought the same.
I meant if you put in the value for a random 2nd, isn't a random 2nd worth a lot more in a 10 team league than it is in a 16 team league? The value range in a 10 team league is from 11-20 overall but in a 16 team league, it's 17-32 overall.
10 Team PPR
Roster: 26 Players + 2 TAXI + 2 IR
Start: 1 QB, 2 RB, 2 WR, 1 TE, 2 Flex (RB/WR/TE)

QB: Aaron Rodgers, Matthew Stafford, Tom Brady
RB: Le'Veon Bell, Aaron Jones, Chris Carson, Marlon Mack, James White, Rashaad Penny, Nyheim Hines, Rex Burkhead, Jerick McKinnon (IR), Qadree Ollison, Rodney Anderson (IR)
WR: Michael Thomas, Mike Evans, Odell Beckham, D.J. Moore, Brandin Cooks, Robert Woods, Corey Davis, Andy Isabella, Tre'Quan Smith
TE: O.J. Howard, Noah Fant, David Njoku, Jonnu Smith, Kyle Rudolph
TAXI: Daniel Jones

Draft Picks:

2020: 2X 3rd, 2X 4th
2021: 2nd, 5X 3rd, 7X 4th
2022: 1st, 2X 2nd

User avatar
UnitedLoon
Role Player
Role Player
Posts: 397
Joined: Thu Apr 12, 2012 8:24 pm
Location: Hudson, Wisconsin

Re: The DLF Dynasty Trade Analyzer

Postby UnitedLoon » Sat Nov 02, 2019 12:16 pm

One thing I just noticed, is that when going from 1QB to 2QB there are some odd player value changes. Njoku goes from 128.4 down to 108.6, where as Herndon goes from 109.9 up to 110.2. I don't get why Njoku is suddenly worth so much less, yet Herndon increases? Don't get me wrong, I know it's just a tool meant for fun and not meant to be the gospel on dynasty trades. Still, it's odd that there is such a drastic difference when only changing from 1QB to 2QB.

User avatar
dlf_kenm
Administrator
Posts: 1274
Joined: Sun May 14, 2006 12:28 am
Contact:

Re: The DLF Dynasty Trade Analyzer

Postby dlf_kenm » Sat Nov 02, 2019 8:09 pm

Hey loon... I think I may have said this before, but awesome logo / avi ! Love it!

Anyway, to answer your question, since the TA draws it's data from different sources this can lead to some surprising differences in player value between regular and SF leagues, For instance, the regular player values are pulled and combined from regular dynasty ADP, our overall dynasty rankings, and trades that happen in real world 1QB leagues.

The values for the 2QB / SF trades are drawn from a different data set - our 2QB / Superflex rankings, our SF ADP, and trades in SF leagues. Since these are different people and different data, these curious variations can occur.

There are as we all know pros and cons to every method. While the upside of constantly drawing from continuously updated data is a big plus in our opinion, it can lead to some weird results.

Anyway, hope that helps. Always welcome and appreciate the feedback!

KM


Who is online

Users browsing this forum: No registered users and 1 guest