Importance of Scouting for ALL TEAMS

[kjohnson]

Send your team captain to alliance selection with something to write with so they can keep up with who has already been picked (or just turn around and look at the screen). Second only to picking a team not at your event, nothing is more embarrassing than choosing a team who is already a member of another alliance.

[Siri]

Quick tip for anyone who doesn't have the resources (people/manhours, passion, something) to do major scouting at events. There's no replacement for good match scouting, but for later events, consider pre-event scouting. I try to at least get full OPR (overall, hybrid, teleop, bridge, CCWM, etc) data for all the teams at my event who've played before, and often watch many match videos, in addition to our scouting. Realize that teams do change a lot between competitions though, especially their first and second (and districts with 6hr unbag time), though the videos can help you get a better sense of who.

Also, even if you only have 1 dedicated person, coach them on what to look for (dunkers, defense, special hybrid and balancing/etc) and talk to them regularly. It's not as strong as objective data (unless of course that person really knows their game), but it's something.

After you struggle through an event or two doing this--setup scouting!

[marccenter]

Scouting of course is valuable for ALL TEAMS. One major challenge for a rookie or sophomore team coach/mentor is help the students understand the importance of scouting when they are a very small team (4-6 active students) who are happy to get their robot to function (spending all their valuable time/talent to make just that happen) and they have never been in an alliance captain position to select others. This was the position we finally found sophomore, FRC Team 3548 in at the Waterford district tournament this year. I have been on other rookie teams where the scouting students determined that their efforts appeared to be in vain (why scout when we will never be an alliance captain?) and lost any interest in gathering meaningful scouting information for the team.

It was also challenging because being the first tournament for many teams, the databases on Chief Delphi (3548) are not populated with historic data before the event, thus scouting during the tournament event itself becomes more important.

Thanks to Dr Ed Law of the Bionic Barons and others who put effort into making the data available for smaller teams - we will be using this data for our upcoming MI District Event at Troy along with the Michigan District rankings sheet on the FIRST in Michigan website.

[Clinton Bolinger]

This year FIRST gives us a lot of data that can be used to make smarter selections even if you don't have a full time scouting team.

By viewing the data from the event standing page (http://frclinks.com/e/r/dt) it can show who (most likely) has the best Hybrid, Bridge, or Teleop.

Just copy the data from the web site and past into Excel. Then sort by the appropriate column that you want to look at.

Personally, I adjust the point values given to the CP (CP*10=Adj. CP). To get an overall performance ranking for teams use the following equation: HP+BP+TP+Adj. CP = Total Points.

This data can also be useful to pre-scout the teams attending your next event.

I have attached the Detroit Data as an example.

Total Points:

Team	Sum of Points
51	542
1023	514
469	472
314	470
326	440
217	369
3538	362
3547	342
453	340
1504	334
2591	333
123	329
2960	324
201	315
1701	307
2224	304
815	303
2620	294
2611	276
3096	275
280	273
3175	269
313	265
1189	260
818	257
519	239
2832	237
2676	227
3115	219
3632	219
4380	215
3619	208
3605	194
3002	193
322	187
2673	180
3769	166
94	142
4002	126
3706	91

Hybrid Points:

Team	HP
51	134
1023	127
469	121
1504	116
314	111
201	110
453	106
123	95
217	93
2960	88
2591	86
326	80
818	73
3619	70
2620	67
2611	65
3002	65
815	63
2224	62
3769	62
519	61
3538	59
1701	59
2676	57
94	57
280	56
3547	55
3175	53
1189	52
2832	52
2673	51
3632	46
313	40
4380	40
322	40
3115	39
3096	38
4002	36
3706	22
3605	6

Teleop Points:

Team	TP
469	231
1023	207
51	168
326	130
123	124
314	119
1504	118
3605	118
2620	117
2960	116
201	115
815	100
2591	97
3538	93
3619	88
280	87
3547	87
217	86
313	85
2676	80
519	78
3096	77
4380	75
453	74
1189	68
94	65
3632	63
2224	62
3115	60
2673	59
1701	58
3175	56
2832	55
2611	51
4002	50
322	47
818	44
3002	38
3769	34
3706	19

Bridge Points (Sorted by Sum of Bridge -> Adj. CP -> BP):

Team	BP	Adj. CP	Sum of Bridge
51	60	180	240
314	80	160	240
326	70	160	230
3538	90	120	210
3547	80	120	200
217	50	140	190
1701	70	120	190
1023	60	120	180
2224	100	80	180
3096	40	120	160
453	60	100	160
3175	60	100	160
2611	60	100	160
2591	10	140	150
818	20	120	140
313	40	100	140
1189	40	100	140
815	60	80	140
2832	50	80	130
280	70	60	130
2960	20	100	120
3115	20	100	120
469	40	80	120
3632	10	100	110
2620	30	80	110
123	50	60	110
4380	20	80	100
1504	40	60	100
519	40	60	100
322	40	60	100
2676	10	80	90
201	30	60	90
3002	30	60	90
2673	10	60	70
3605	30	40	70
3769	30	40	70
3619	10	40	50
3706	10	40	50
4002	20	20	40
94	20	0	20

[Tom Line]

We make scouting mandatory for every member of the team. There is a schedule, and they fill out a questionaire as the team plays on the field. It's fairly simple. With a smart scouting sheet, you really only need 2 students scouting each match.

Sure, some students don't want to. I'm sure many of the athletes in sports don't like working out all the time either, but they'll never win unless they do.

[Grim Tuesday]

Originally Posted by Tom Line

We make scouting mandatory for every member of the team. There is a schedule, and they fill out a questionaire as the team plays on the field. It's fairly simple. With a smart scouting sheet, you really only need 2 students scouting each match.

Sure, some students don't want to. I'm sure many of the athletes in sports don't like working out all the time either, but they'll never win unless they do.

The easy part is filling out sheets. The hard part is making decisions about an alliance based on those sheets.

[Tom Line]

We rarely if ever have problems with the decision portion. It's all driven by weightings in an excel spread sheet, and if there are any that are too close to call a quick check of OPR gives us a good tie breaker.

[Debbie]

Originally Posted by Clinton Bolinger

This year FIRST gives us a lot of data that can be used to make smarter selections even if you don't have a full time scouting team.

By viewing the data from the event standing page (http://frclinks.com/e/r/dt) it can show who (most likely) has the best Hybrid, Bridge, or Teleop.

Just copy the data from the web site and past into Excel. Then sort by the appropriate column that you want to look at.

Personally, I adjust the point values given to the CP (CP*10=Adj. CP). To get an overall performance ranking for teams use the following equation: HP+BP+TP+Adj. CP = Total Points.

This data can also be useful to pre-scout the teams attending your next event.

I have attached the Detroit Data as an example.
[/code]

So you are ranking teams based on the alliance score in this example? If you are using alliance stats, you aren't getting a true picture of individual robots. Your posted ranks vary greatly from our individually scouted ranks. Some teams are carried by alliances, and some are hurt by alliances, which is why individual scouting is needed. (IMHO)

[Clinton Bolinger]

Originally Posted by Debbie

So you are ranking teams based on the alliance score in this example? If you are using alliance stats, you aren't getting a true picture of individual robots. Your posted ranks vary greatly from our individually scouted ranks. Some teams are carried by alliances, and some are hurt by alliances, which is why individual scouting is needed. (IMHO)

Absolutely ... trust me our team does individual scouting at the events we attend. However, the method described above is a simple way to look at events you are not at. This is also helpful when forecasting or looking forward to the next event you attend.

Also I feel that the first few picks of the draft can be determined by the above ranking. The first pick is easy, but the 24th pick is the hardest pick of the draft.

This was just a suggestion for teams that don't have the resources or man power to do full comprehensive scouting system.

If anyone is interested in how we scout at an event feel free to stop by our pits and ask.

-Clinton-

[Debbie]

Originally Posted by Clinton Bolinger

Absolutely ... trust me our team does individual scouting at the events we attend. However, the method described above is a simple way to look at events you are not at. This is also helpful when forecasting or looking forward to the next event you attend.

Also I feel that the first few picks of the draft can be determined by the above ranking. The first pick is easy, but the 24th pick is the hardest pick of the draft.

This was just a suggestion for teams that don't have the resources or man power to do full comprehensive scouting system.

If anyone is interested in how we scout at an event feel free to stop by our pits and ask.

-Clinton-

Ahh ok.. because this is certainly not the order we picked in, but I guess if you are looking at events you haven't been at, it can be helpful. It can also be a little misleading in that there are still teams that are floating up there that you would not pick as a top pick (and some down low that you would want to pick) but if a team isn't doing individual scouting, then I guess that would help! I figured you guys were scouting individual robots, just was wondering when you posted that because we were there and it does not match our data.

[Ben Martin]

I agree, having a strong knowledge of the teams at an event for draft-list making and match preparation is extremely important. I've had this mindset ever since a mentor in 2006 yelled at us for giving him bad scouting data. We buckled down and got better, and the match results improved as the scouting data improved.

At our last event, for making a pick list, we ordered the teams by coopertition bridge-adjusted OPR and cross referenced it to our scouted data.

We also like to print papers with our scouting comments and averaged data on it for theoretical 3-vs-3 match-ups. This helps when we discuss strategy during qualifications as a way to show that we know what we're doing.

[junefish]

Originally Posted by Tom Line

With a smart scouting sheet, you really only need 2 students scouting each match.

Respectfully, I disagree. Especially given this year's game--that, theoretically, there can be an infinite number of balls scored, rather than just filling up a rack, like last year--it's very important to have each robot in a match given individual attention. If one person scouts just 2 robots at once the accuracy goes down about 40%--more if the 2 robots are high scorers. (In LA we had 6 people, and we still only got about 60% of the total possible data! ) It may be difficult to have 6 scouts in the stands at all times, but it really pays off during the picklist creation.

[junefish]

Originally Posted by pfreivald

Scouting is even more critical than just making good picks (or avoiding embarrassing mistakes) during alliance selection -- it also helps you get there.

Before each match, your drive teams (or at least drive coaches) should be planning what each robot is expected to do -- that includes both alliance partners and opponents. The more you know about your opponents, the better you can plan how to handle them... and the more you know about your allies, the better you can make your plan.

Especially early in qualifications, teams tend to tell you what their robot was designed to do, not what it actually can do. As games pass and the evidence stacks up, teams tend to be more honest -- though you often get the "but it's fixed now... we think" line (which sometimes turns out and sometimes doesn't). Sometimes you'll be surprised when opponents suddenly come out with a new capability (because something was fixed or upgraded or what-have-you between matches). But understanding what each robot is capable of, and how it manages to accomplish those tasks, is very important to making plans before each and every qualification and elimination game.

So for example, if you know from your scouting data that one of your partners (or your own robot) can't score, and that two of your opponents score only from the fender, it makes perfect sense to send the non-scoring robot to the other side to play defense -- even if they can't stop all the scoring, they can likely slow them down, and this might be enough to give your alliance the edge. If your opponents are strong shooters or incapable of scoring altogether, maybe defense is not the best option and they should spend the entire game doing bridge-related tasks or feeding balls to your good scorers or playing counter-defense instead.

I completely agree--our drive team was really helped by the information that the scouts provided them, but we were still hampered by not having complete, accurate data on all teams. I honestly do not think we would have gone as far as we did in Los Angeles (Elimination--Finals) if we had not been able to use our data to predict the opposing alliances' strategies.

[gyroscopeRaptor]

Pit scouting is useful for all teams, whether you have a box-bot or a champion machine. We learned this firsthand. Last year, we had no scouting beyond notes.

This year, we made a form looking for details on all robots and scouted them all by 10 on Friday (62 team event, one group of 2 pit scouts) and had paper dossiers on each robot. Since our plan was mainly balancing, we learned after our second match (a 120 lb robot trying to balance with a 90 lb robot) that we needed to plan based on weight, form factor, and CoG to find the best alliance/co-op balance partner. We lined up perfectly in these aspects with the top teams in our event, and we would never have known this if it wasn't for pit scouting! We ended up balancing with the best.

[vikesrock777]

As the coach for Apple Pi through two events, I don't think I can stress the importance of scouting enough. It truly is important regardless of your team ranking, and I prefer to have some data before the first qualification match even starts.

At WPI, a smaller regional, it was often very rushed to create the match strategy for the next match. Any and all data I had from my scouting team helped expedite this process massively. Without a solid scouting team, I don't think we would have done so well and ended up seeded 2nd. As a picking alliance, the data was massively helpful. Without the data, I don't think we would have been able to win WPI (thanks 3205 and 190!)

At Connecticut regional, we had our pit scouting almost entirely filled out by the end of Thursday. Walking in Friday morning I already had a notebook filled out with a page for each match. I had written down each team and some of their basic information as given by pit scouting. I also included OPR's from Ed's thread (thanks so much for the data!) for teams that had already competed to aid in giving an evaluation of how the team was going to compete. Using this I was able to better plan for what we were up against every time. It didn't always work out the best in the end (we seeded something like 30th due to a very rough qualification schedule), but we always had a strategy going into the match. Going into Saturday, we had all of our match data compiled, which helped even more. We knew approximately how well each robot was performing which helped in the remaining qualifications. Going into elims, that data proved even more useful. We were able to use our data to very accurately predict our opponent's strategies, and make good counter-strategies. With this data we made it to the finals (thanks 558 and 1071!) Without this data, I don't think we would have done as well.

In summary, scouting helps all throughout the event. It is not exclusively for teams who are going to be in picking position. All teams can benefit from it throughout the event. It can help you in every qualification match. You never know, it may even aid you into being in a picking position, or it may be crucially helpful to your elimination alliance!