Sykes Scouting Database 2020

This is a scouting database which provides Elos, rates, calculated contributions (OPRs), and iterative logistic strengths for all teams at each event using the data from the TBA API. Each sheet contains data from a distinct FRC event. A new database will be published weekly within a day or two of all of that week’s events being completed. For sheets which contain events that have not yet occurred, seed values for each category are available in order to aid in pre-scouting.

You can download it from Github here

Here are the threads from the previous years’ scouting databases for the interested:
2016
2017
2018
2019

The previous season databases are also on the same linked GitHub page under the archives folder.

27 Likes

Here is the direct download for the initial release.

Well we’re kicking off the 2020 season very shortly with the Israel competition. There’s not a ton in this first book as no teams have played in any official events yet. All seed values except Elos and ILSs are set to averages from the initial week 0 competition. Elos are set to the start of season ratings as described in my Elo thread. ILSs are based off of the RP rates in week 0 (none, so I used 1/29 for both) and adjusted based on Elos.

Not many major changes yet, the biggest one will be the addition of Zebra Motionworks data for equipped events. I have added a few more details to the “Event Summaries” sheet since last season.

As always let me know if you have suggestions or bug reports. Early weeks tend to have more bugs, and if you have ideas for metrics that I don’t have listed, it’s much easier for me to add them now than a few weeks into the season.

4 Likes

We’ve got a small first update. Here is v0.3

I just changed some of the formatting in the “Event Summaries” sheet. Also by switching the function used for the hyperlinks in that sheet, the hyperlinks are now functional on older excel versions.

1 Like

Such an amazing resource! Will you be updating your Event Simulator as well?

3 Likes

I better! It’s inextricably linked to the scouting database. :smiley:

The early Israel events took away my standard buffer period between week 0 and 1, but at least they will provide much better testing data than week 0 :man_shrugging:. My goal is to have an initial upload by Wednesday with basic functionality, and then I’ll need a few more days to get up to the full feature set available last season.

Every year so far has had a smoother initial update process than the prior year because I improve the update tools, code structure, and comments as I go. Unfortunately I feel like it’s always about the same amount of total work each season since the scope keeps growing so I have more to update :grinning:.

My big project this season has been to make a guide for how to do the full start of season update. The hope with that is that I won’t waste time doing things out of order in future seasons. I also would like to have a guide like this if/when I lose my drive for this work so that someone else can pick up for me if they want.

3 Likes

This looks great ! thank you !!!

Do we need to setup the TBA API in our copy of the worksheet? How does it update throughout the event?

Thank you for putting this together!

This book has no live update capabilities, I’ll release updates each week which contain data from all events that have played up to that point in time.

I’ll be putting out a separate book called The “Sykes Event Simulator” in the next day or two that does have live update capability.

2 Likes

Updated to v1.4: https://github.com/inkling16/SykesScoutingDatabase/raw/master/Sykes_Scouting_Database_2020.1.4.xlsx

Here’s the big week 1 update. I added much better seed values, as well as populated data from all completed week 1 competitions. I also added 3 new columns to the event summaries tab: Attending teams, has Zebra MotionWorks, and isCompleted.

Let me know if anything looks off, it’s particularly tricky to get the formatting right everywhere, so it wouldn’t surprise me if I mislabeled a header or something on one of the sheets.

1 Like

Here are the Pareto-optimal week 1 performances if we break the game down into auto, teleop exc. endgame, and endgame:

team Number Nickname Event auto Points teleop Excluding Endgame Points endgame Points
694 StuyPulse Palmetto Regional 28.16 32.25 30.94
973 Greybots Los Angeles North Regional 22.97 46.85 32.52
1540 Flaming Chickens PNW District Clackamas Academy Event 26.25 33.47 18.41
1619 Up-A-Creek Robotics Great Northern Regional 30.23 29.06 17.56
1678 Citrus Circuits Los Angeles North Regional 30.33 27.15 13.63
1736 Robot Casserole Miami Valley Regional 20.39 21.20 35.95
2337 EngiNERDs FIM District Kettering University Event #1 23.97 14.51 31.97
2910 Jack in the Bot PNW District Glacier Peak Event 23.63 32.73 23.52
3824 HVA RoHAWKtics Palmetto Regional 14.47 9.81 38.55
5172 Gators Great Northern Regional 28.68 47.40 17.34
5413 Stellar Robotics Miami Valley Regional 23.62 23.30 31.28
6560 Charging Champions Los Angeles North Regional 34.29 19.91 19.54

Notable omission in 1690, who had the highest week 1 Elo. Their performance in these 3 metrics was dominated by 973 though.

4 Likes

Also, here’s the official top 25 Elos for the week:

team Number Nickname winning Margin Elo
1690 Orbit 1842
2910 Jack in the Bot 1830
1678 Citrus Circuits 1828
973 Greybots 1823
5172 Gators 1811
3538 RoboJackets 1795
1540 Flaming Chickens 1788
1619 Up-A-Creek Robotics 1785
694 StuyPulse 1781
1986 Team Titanium 1778
2471 Team Mean Machine 1777
2337 EngiNERDs 1770
3674 CloverBots 1767
365 Miracle Workerz 1767
1918 NC GEARS 1758
1519 Mechanical Mayhem 1751
133 B.E.R.T. 1750
2468 Team Appreciate 1746
115 MVRT 1742
1807 Redbird Robotics 1736
2200 MMRambotics 1734
1425 Error Code Xero 1732
33 Killer Bees 1730
188 Blizzard 1729
610 Crescent Coyotes 1728
4911 CyberKnights 1728
340 G.R.R. (Greater Rochester Robotics) 1725
3641 The Flying Toasters 1725

These do differ a bit from the rankings shown on FUN last night due to the calculations now using the stdev of week 1 WMs instead of week 0 WMs.

4 Likes

image YAY!

1 Like

Maybe I’m just missing it somewhere but what stdev are you using for WM’s?

I’m using 58.3 as the stdev of all week 1 qual and playoff matches, it’s hidden in the code for my event simulator.

1 Like

The numbers reflected in your AWESOME Scouting Database Excel file, in the World Results tab…are those calculated based on qualification matches or qm’s plus the elimination rounds?

Calculated Contributions/OPRs are quals only
ILSs are quals only
Rates are quals + playoffs
Elo is quals + playoffs but playoff matches have 1/4 the rating impact of quals matches

1 Like

Heck yeah #paretooptimalgang

1 Like

Are you keeping an updated list of team Elos somewhere throughout the season?

Yes, in the “Seed Values” sheet of this scouting database.

1 Like

Updated for week 2: https://github.com/inkling16/SykesScoutingDatabase/raw/master/Sykes_Scouting_Database_2020.2.0.xlsx

I currently do not have West Valley included. For some reason, instead of just giving the 1 seed a bye in the quarterfinals, the FMS seems to have required the 1 seed to win a dummy match against teams that did not even attend the event. I have no easy way to handle this strange case right now, so I’ll need some time to sort out this event.

14 Likes

If I could click “Like” 100 times, I would. Thank you for a quick turnaround on Week 2!!