**2017 Weeks 8&9 (Houston & St Louis) raw score data
2017 Weeks 8&9 rotors1234 averages**
**
if links are broken, look for posts in this thread for updated weeks 8 & 9 links
What else do you want to see?**
Links to 2017 Weeks 1 thru 7 data and analytics here.
What else do you want to see?
2017 Weeks 1thru9 Qual & Playoff match component Score Data*
All qual and playoff match component score data for all events weeks 1 thru 9 in one large sheet.
All Booleans converted to 1 and 0 for easier summing/searching.
“Event Name” Column CC added for easier sorting by event.
***** if link is broken, look for posts in this thread for updated weeks 8 & 9 links
*
*
Thanks!
If you are taking requests, I would really like to see a count of which tiebreakers were used throughout the season.
Tiebreaker Used Count
Fouls 9
Auto 4
Rotors 0
Climbs 0
Pressure 1
Replayed 43
Thank you
2017 entire season (Weeks 1thru9) qual match Component Event OPR*](https://www.chiefdelphi.com/media/papers/download/5077)
All 160 events from the 2017 season which included qual matches.
OPR calculated for each event individually (aka “Event OPR”).
OPR calculated for all qual match score components available from TBA (thanks Eugene and Phil), even ones whose OPR usefulness is questionable.
For purposes of this calculation, Booleans are treated as 1 (true) or zero (false).
A separate file is provided for each event, in case you are interested in just a few events.
Addtionally, a large file containing all the data in the separate files is provided. This large file contains two extra columns: eventName and Week#, so you can use those columns as sort keys.
***** if link is broken, look for posts in this thread for updated links
*
*
You’re amazing Ether. Thanks again.
I would like to learn how many matches were won with a 0 vs 1 vs 2 rotor auto.
Basically to examine how important it was to build a side gear auto function.
Dave
For qual matches only, weeks 1 thru 9:
blue won r2, r1, r0:
20/23=86.9565% 3648/6044=60.3574% 2665/6614=40.2933%
red won r2, r1, r0:
19/23=82.6087% 3456/5999=57.6096% 2658/6659=39.9159%
*For all you stats mavens out there.
Here’s something I haven’t seen posted anywhere yet:
Event Component OPR Residual Descriptive Statistics](https://www.chiefdelphi.com/media/papers/download/5087)
Descriptive Statistics (minimum, first quartile, median, third quartile, maximum, mean, standard deviation, skewness, and kurtosis) of the qual match Component OPR residuals for each Event in 2017.
If anyone wants to play with the raw residuals (e.g. plot them), I can post that too.
If there are any students interested in further explanation please post here or PM me.
Here’s](https://www.chiefdelphi.com/media/papers/download/5088) a summary of the descriptive stats for the residuals of the Event OPRs for total points, showing InterQuartile range, skewness, and kurtosis.
Some folks are wondering “what’s a residual?”
Say you have 3 teams X, Y, and Z on an alliance, and in qual match N at event E that alliance’s final score is 400.
Let’s say teams X, Y, and Z have a final score Event OPR of 100, 150, and 120 respectively, for that Event.
The residual for that alliance for that match would be 400 - (100+150+120) = 30.
So that residual gives you an idea how well (or poorly) the OPR model fit that data point (final score).
If you consider all the residuals for an event, it will be a vector with 2*M elements, where M is the number of qual matches played at that event.
That vector of residuals gives you an idea how well (or poorly) the OPR model fit all the final scores at that event.
You can then look at the descriptive statistics for that vector of residuals.
Sweet! Thanks Ether.
Dave
In case anyone is wondering, here’s the AWK script I used:
*
*$2=="qm" {
if($18>0)b2++
else if($9>0)b1++
else b0++
if($52>0)r2++
else if($43>0)r1++
else r0++
if($72>$76){ # blue won
if($18>0)bw2++
else if($9>0)bw1++
else bw0++
}
if($76>$72){ # red won
if($52>0)rw2++
else if($43>0)rw1++
else rw0++
}
}
… using this](https://www.chiefdelphi.com/forums/showpost.php?p=1681899&postcount=3) raw data.
**I don’t think I ever posted all these links in one location. So here they are in case anyone is searching for them:
2017 Weeks 1thru9 Q&P raw Score Data
2017 Weeks 1thru9 qual match Component Event OPR
Event Component OPR Residual Descriptive Statistics
Event OPR TotalPoints Residuals Stats.zip
2017 weeks 1 thru 9 Qual & Playoff raw component data
2017 World Component OPR weeks 1thru9 qual match
2017 Team Event Component OPR Averages weeks 1thru9 qual**
