+- International Simulation Football League (https://forums.sim-football.com)
+-- Forum: Community (https://forums.sim-football.com/forumdisplay.php?fid=5)
+--- Forum: Media (https://forums.sim-football.com/forumdisplay.php?fid=37)
+---- Forum: Graded Articles (https://forums.sim-football.com/forumdisplay.php?fid=38)
+---- Thread: *Expected Value of Draft Picks 1-24 (/showthread.php?tid=7230)
*Expected Value of Draft Picks 1-24 - Molarpistols - 01-28-2018
Anyone who has followed just about any sport and the drafting strategies involved has doubtlessly happened upon a "Draft Pick Trade Value Chart" of some sort. There are many examples a short Google away, including NFL DRAFT TRADE VALUE CHART, NHL DRAFT VALUE CHART. The purpose of these charts is to give a better understanding on the value of a draft pick in terms of trading power. It stands to reason that a pick closer to the front of the draft is worth more than one near the end, as you can have your pick of the litter so to speak.
I started off this project intending to make something similar for the NSFL, using past drafts as an indicator. I'm not a rookie GM any longer, but I had wished for something like this when I first started out. The NSFL is my first and only sim-league, so it has been difficult at times for me to judge the value of these picks. It's fairly easy to judge a player's worth, we have a number to go by, TPE. You can also make judgements based on their activity level and personality from their time on the forums and the Discord. But how do you judge a draft pick? How are you to know if that player is going to retire after one season, go inactive fairly soon after being the first EVER draft pick in the NSFL, or if they're going to spend four seasons languishing in the DSFL? To make those judgements, I chatted with my team staff, but at least on my end, it pretty much came down to a gut feeling. I enjoy having statistics and numbers to back up my reasoning, so I set out to do just that.
I'll talk about my process leading up to the findings for a little bit. First, I talked to @timeconsumer to find useful Excel functions that I could use to delve into my data. The Excel genius that he is, he introduced me to handy tools such as INDEX and MATCH, which when used together can output lots of useful information for you.
I used the INDEX-MATCHing tool I constructed to calculate the total TPE for all the base archetypes. I figured it'd be a good way to test to ensure it was working properly, and I could use that information to determine each prospects total TPE earned. The player pages in the team rosters have a figure for TPE earned, but I've noticed (at least in the past, it could be fixed now) that they aren't always 100% accurate. Once I started diving into each individual prospect, I quickly scrapped the idea of using their archetype TPE to determine how much they had earned. With all the position swaps and archetype changes it'd be a bit of a mess to sort it all out in a timely fashion while still being able to glean useful information from it.
So instead of using each individual archetype TPE, I averaged all of the archetype TPE starting points, which came out to 436.1111. This does not include the initial 50 TPE a starting player can add to that, it's just that the average archetype starts at 436.1111. I made the assumption that position/archetype aren't a huge reason why a specific player gets picked higher in the draft, an active player will go really early. It doesn't matter if their particular archetype or position starts at a slightly lower TPE, so the average starting TPE would work just fine for my purposes.
For anyone who has been around awhile, the NSFL draft has been a bit of a crapshoot when finding how many draftees will be available each round. The first two drafts (Season One and Season Two) were quite large, having double digit round numbers in the end, with several late round gems being found. But it would be awfully silly to try to determine the value of an 18th round pick when there's been one that stayed active (I'm just guessing, there might be several 18th round picks that stuck around, but no 19th rounders that did, etc etc..) Season Four's draft had all of one round of active prospects and Season Five had somewhere around 2-3 rounds worth.
I chose to value the first 24 picks, mostly based on a gut feeling that we've been pretty good about getting pretty close to 20 prospects each draft, be it new recruits or re-creates. This covered the first four rounds of the season 1 draft, where we had 6 total teams and the first three rounds of season 2 and 3. I again used my gut to decide that I'd base the initial value off the first three seasons of drafting, first to exclude the Season Four draft with like 10 prospects, and second, to give 3 seasons of TPE earning as a decent baseline for value. The maximum contract for a rookie is three seasons, and my completely anecdotal evidence suggests that almost all rookies sign for three seasons.
The first graph will show the Average TPE earned at the end of the prospect's first season, with pick 1-24 along the x-axis.
![[Image: HVeUAB0.jpg]](https://i.imgur.com/HVeUAB0.jpg)
The graph shows pretty much what everyone was expecting, the higher the pick, the higher the TPE the player will generally accrue by the end of their first season. As the picks progress, the TPE earnings also progressively dip. Up until about pick 12, one can reasonably assume their pick will have about 200 TPE at the end of the first season. There will obviously be some variance, but we're basing it off three players, a single outlier and the number will be skewed horrendously at this point. Picks 13-24 you can reasonably expect 100-150 TPE being earned, unless you picked Khalifa Al-Mahrooqi at #20 and he retired almost immediately afterward...
This second graph will show the average TPE earned at the end of the prospects second season, again based by pick number.
![[Image: QShCCk3.png]](https://i.imgur.com/QShCCk3.png)
This graph shows a similar trend, higher picks earn more TPE, who would've thunk it?!? Pick #20 is cursed. Al-Mahrooqi retired at the end of one season, and Masdival, also a #20 pick was DSFL'd or went inactive or something, I can't really remember. The top 12ish picks can be expected to be around 300-400 TPE at the end of their second season, an increase in 100-200 TPE from their first season. 13-24 can be expected to earn about 50-100 TPE in their second season.
And the final graph, shows the same stuff but at the end of the prospect's third season.
![[Image: GmqAW2k.jpg]](https://i.imgur.com/GmqAW2k.jpg)
The overall trend continues, the first twelve picks are all pretty similar in value based off of TPE. 13-24 sees a lot more variance. There are a few good picks that brought averages to high peaks, but then we still have Al-Mahrooqi, Masdival and Rico Rodriguez stagnating the heck out of pick #20.
Moral of the story: DON'T EVER MAKE A PICK AT #20!
Just kidding, a bust can happen at any pick. Look no further than Shaka, the first ever pick in the NSFL that went inactive after a season. The average for that pick was brought up mightily by our almighty dictator, Dermott, with some help from a worthy GM, Antonio Sandoval. You can also strike gold later in the drafts, with picks like Jayce Tuck, Owen Taylor and Wallace Stone being picked at #18.
What did we learn? Not much. How much time did Molarpistols waste finding nothing? More than I had hoped. Anyway, thanks for sticking around!
Code:
1289 wordsGRADED
*Expected Value of Draft Picks 1-24 - Evok - 01-28-2018
Lets go #18 !
*Expected Value of Draft Picks 1-24 - Raven - 01-28-2018
77th overall or bust
*Expected Value of Draft Picks 1-24 - timeconsumer - 01-28-2018
Notice how you stop at #24 when I was drafted 25th overall? What's the matter? I'm screwing up your narrative?