I like to play with spreadsheets and thought I’d take a look at Regional field performance numbers. So far FIRST seems to be meeting it’s average goal of 10 matches per event. Large events have fewer matches, while small events have more.
The worst so far has been the Israel turnaround (~13 minutes/match on average) with all of their unique communications problems.
The best so far has been the Oregon Regional at a few seconds under 7 minutes per match.
Anyone think of other interesting statistics to track on an event basis?
I remember seeing Seattle and Oregon on the top of a simular turnaround list last year. It seems like the Pacific Northwest has some great volunteers that really know how to work the field
Coincidentally, I’ve been doing the same thing today.
I’ve broken down the turnaround times by day section on the qualification matches. For a normal regional, session 1 is Friday morning, session 2 is Friday afternoon, and session 3 is Saturday morning.
Week-Event------------------------Session 1--Session 2--Session 3--Overall
1----Autodesk Oregon Regional-----07:27.50---06:58.29---06:36.00---06:58.65
1----Greater Kansas City Regional-07:24.00---07:12.86---06:45.52---07:07.50
1----Peachtree Regiona------------07:56.47---07:30.00---06:10.00---07:11.39
2----Chesapeake Regional----------07:45.71---06:50.00---07:25.38---07:13.48
2----Wisconsin Regional-----------08:22.86---06:51.67---06:47.50---07:14.07
2----Arizona Regional-------------10:06.32---07:36.00---05:45.00---07:28.35
1----BAE Systems Granite State----08:51.43---07:05.63---07:17.50---07:38.18
1----New Jersey Regional----------10:00.00---07:20.00---07:01.71---07:38.18
1----Finger Lakes Regional--------07:27.00---08:33.10---06:46.36---07:41.41
1----Washington DC Regional-------10:03.75---07:38.05---06:45.52---07:47.44
2----Florida Regional-------------09:06.32---06:38.05---08:41.54---07:48.14
2----WPI Regional-----------------07:36.67---08:02.86---07:48.75---07:50.18
2----Pittsburgh Regional----------09:30.00---07:35.56---07:53.08---07:54.92
2----San Diego Regional-----------09:40.00---07:24.32---07:32.50---07:57.72
2----FIM Ann Arbor----------------10:21.82---08:04.62---08:11.11---08:26.49
2----New York City Regional-------10:28.42---08:50.63---08:20.00---09:02.22
1----Bayou Regional---------------10:55.71---08:27.86---08:26.25---09:03.10
1----FIM Traverse City------------10:25.00---09:37.50---08:06.21---09:09.04
2----FIM Cass Tech----------------10:33.33---09:43.64---09:06.67---09:35.65
1----FIM Kettering University-----13:30.00---09:34.55---09:22.50---09:49.01
2----Israel Regional--------------21:24.00---15:00.00--------------17:22.22
X----Overall Average--------------09:29.40---08:03.60---07:31.20---08:12.00
X----Average (without Israel)-----09:04.80---07:49.20---07:31.20---07:59.40
You can see the improvement through many of the events. Arizona managed to get to a blistering pace of 5:45.00 by Saturday morning.
Looking at this has also changed my opinion of Kansas City for the better. As an attendee and volunteer, it felt like we were way behind all day Friday. When you look at the statistics like this, we were attempting more matches than any other regional so far and maintaining a very good match turnaround time relative to the rest of the country. :]
Starting six matches behind can really make you move quickly… Arizona’s Session 3 is including a double-length field reset due to the field router going down suddenly.
Arizona started slow and gained pace in the afternoon. About 3:15, the inspectors finally finished, so I wandered out to watch a few matches. I spent the rest of the afternoon following the FTA around with a stopwatch to tell him when extra speed was needed in a reset. Did that Saturday morning, too, and the speeds went really high–one of the queuers told me there was a 5-minute turnaround. Then I had to go inspect some more (we started elims inspections at something like 10:30-11:00).
The AZ inspectors were really thinking ahead. They knew that we had been to San Diego, so they were in our pit as soon as the robot was unpacked and on the cart, asking if we were ready to inspect. Saturday, they had us inspected for elimination, before the last matches were completed. Thanks to all the volunteers that make the competitions happen.
Silicon Valley & St. Louis both beat the best average time set by Oregon.
West Michigan seemed to be very regimented and liked to start matches exactly on the minute mark. they stayed almost right on the schedule pace they set for themselves.
The Virginia Regional ran 95 matches, with 9 matches per team. The field cycle was really odd compared to the NJ Regional, going about 3 minutes slower due to problems that went unexplained. I think it had something to do with WPA keys and and router models. The FMS live-scoring died in the middle of the marathon semifinals, which left me a little edgy the entire time. All I know is that we never broke for lunch, which was a little depressing. 422 is not a very large team, and having to video tape every match, I couldn’t leave the center all Saturday.
The regional was lots of fun, I just hoped the match cycle would have gone down a minute, not up three minutes.
Week--Event----------------------Session 1--Session 2--Session 3--Overall
3-----St. Louis Regional---------06:46.36---06:31.11---06:00.00---06:27.76
3-----Silicon Valley Regional----06:57.27---06:34.55---06:43.85---06:43.70
1-----Autodesk Oregon Regional---07:27.50---06:58.29---06:36.00---06:58.65
3-----Dallas Regional------------09:08.57---06:46.67---06:16.80---07:03.20
1-----Greater Kansas City--------07:24.00---07:12.86---06:45.52---07:07.50
1-----Peachtree Regional---------07:56.47---07:30.00---06:10.00---07:11.39
2-----Chesapeake Regional--------07:45.71---06:50.00---07:25.38---07:13.48
2-----Wisconsin Regional---------08:22.86---06:51.67---06:47.50---07:14.07
3-----FIM Western Michigan-------07:00.00---07:41.43---07:00.00---07:22.03
2-----Arizona Regional-----------10:06.32---07:36.00---05:45.00---07:28.35
3-----Boilermaker Regional-------07:24.00---07:30.00---07:50.53---07:33.91
1-----BAE Systems Granite State--08:51.43---07:05.63---07:17.50---07:38.18
1-----New Jersey Regional--------10:00.00---07:20.00---07:01.71---07:38.18
1-----Finger Lakes Regional------07:27.00---08:33.10---06:46.36---07:41.41
3-----Midwest Regional-----------10:10.59---07:01.82---06:45.88---07:45.67
1-----Washington DC Regional-----10:03.75---07:38.05---06:45.52---07:47.44
2-----Florida Regional-----------09:06.32---06:38.05---08:41.54---07:48.14
2-----WPI Regional---------------07:36.67---08:02.86---07:48.75---07:50.18
2-----Pittsburgh Regional--------09:30.00---07:35.56---07:53.08---07:54.92
2-----San Diego Regional --------09:40.00---07:24.32---07:32.50---07:57.72
2-----FIM Ann Arbor--------------10:21.82---08:04.62---08:11.11---08:26.49
3-----Virginia Regional----------07:24.00---08:57.50---08:47.27---08:43.52
3-----FIM Detroit----------------08:55.71---08:15.88---09:31.20---08:49.32
2-----New York City Regional-----10:28.42---08:50.63---08:20.00---09:02.22
1-----Bayou Regional-------------10:55.71---08:27.86---08:26.25---09:03.10
3-----Utah Regional--------------10:52.00---08:18.26---08:21.43---09:03.46
1-----FIM Traverse City----------10:25.00---09:37.50---08:06.21---09:09.04
2-----FIM Cass Tech--------------10:33.33---09:43.64---09:06.67---09:35.65
1-----FIM Kettering University---13:30.00---09:34.55---09:22.50---09:49.01
2-----Israel Regional------------21:24.00---15:00.00--------------17:22.22
X-----Overall--------------------09:29.40---08:03.60---07:31.20---08:12.00
X-----Overall (except Israel)----09:04.80---07:49.20---07:31.20---07:59.40
Western Michigan appears to have gone like clockwork. It’s absolutely uncanny, when compared to the rest of the data. My congratulations to everyone involved.
St. Louis looks to have gotten off to a very fast start, with the current record for Friday morning and Friday afternoon, but no one has approached Arizona’s speed for Saturday mornings yet.
Virginia also ran a couple of matches Saturday morning before what I suspect was a break for the opening ceremonies. I didn’t count the time across opening ceremonies, since that is not remotely under control of the field crew. I saw the same thing on New Jersey when I was doing my initial analysis and handled it the same way. New York might have done that, but there were two morning delays and I couldn’t guess which one to discount (if anyone who was there can tell me, I’ll be happy to update the averages accordingly).
There were a few 30-40 minute stops during the day Friday for unexplained reasons, and we were so far behind by the end of Friday they decided to try and make it up early Saturday, pausing around 9:30 for a quick opening ceremony.
The field problems continued on Saturday. I cannot blame the field staff; they were the ones trying to fix it. Although not as bad as some other regionals, I did notice people who came in off the street to watch were put-off by the delays and walked out. When FMS finally failed and they went to hand-scoring, and then went with only 2 representatives for most awards, I knew they had pulled out all the stops in an effort to minimize the effects of running so very late. (several ties in the semifinals didn’t help either)
I’d love to say “Why can’t FIRST make a more robust FMS”, but that would be doing a disservice to everyone: The FMS is actually quite good, but I think that we’re still stressing it further than it can handle sometimes. I don’t know the architecture (Windows?) but perhaps they will redouble their efforts on robustness next year.
The West Michigan regional was incredibly well-run. I suspect the average would have been even better than it was, but the field shut down for over 40 minutes because someone was running their N router in the pits and it was preventing the teams from connecting to the field, so they stopped running matches.
I’ll let the FTA and other field crew get into as many specifics as they feel are necessary, but rest assured, the field crew was doing absolutely everything they could to get the event running as quickly as possible. There were issues with communications and FMS that slowed the event down dramatically (especially interesting considering the same field had almost no issues with it the previous two weeks in DC and Baltimore).
While I wasn’t volunteering in an official capacity this weekend, and only attended the regional on Saturday, I have worked with most of the volunteers at the event before. They are some of the hardest working and best volunteers I’ve come across, and are typically great at solving field issues (VCU was a week 1 regional until last year).
We’re lucky that VCU is so supportive of FRC, and the delays were taken in stride by the VCU employees (unlike the crews at some other venues).
Our FMS/communication problems were problems with robots linking with FMS because of differences in wireless radios. The problem arose when any robot using the original (square and black) Linksys gaming adapter linked with FMS before a robot using the newer, silver adapter. This caused a problem in FMS and the remaining radios would not link, causing FMS to have to be restarted.
We eventually worked out a process to solve this link problem: We first ensured all driver stations were linked, then allowed robots with new radios to power up and link, then booted all remaining robots with old radios. After we figured it out, this longer and time consuming process worked most of the time, but we still had to restart the FMS router periodically.
We began keeping track of teams using the newer radios, and used this process in each match they competed in. We ended with 7/63 teams using new radios (about 11%). Those 7 teams cycled through about every 10 matches. Time between matches without new radios ran about 6.5 minutes, matches with new radios of course ran longer.
Constantly turning around matches minutes faster than many of the other regionals!