Quote:
Originally Posted by Boltman
Agree, electronic collection offers another avenue of learning. In a way though today's kids are already overloaded with electronics and there is nothing wrong with teaching them to make observations, discuss strategies pre-match and learning to evaluate the entire team while taking relevant notes. In a way, simply getting away from electronics to do everything itself is a learning experience for them. One problem with today society is much of it believes an app solves everything. It doesn't. Many of these kids were born with various electronics in hand and lack the ability to evaluate manually and look beyond stats and at other clues not mentioned much on CD.
I see value in both aspects and one is not better than the other. So when the OP asks "Best way to scout" I don't take that as "best way to learn about statistics and app development" necessarily.
We all have ways that we think works and probably tweak it yearly... they are all learning opportunities. I go back to sample size and number of matches being an issue with heavy reliance on quantitative scouting. Most teams only play about 22 minutes per regional in qualifications and only against and with certain teams so I find that numbers are easily skewed base on pairings in any single event. I try to cut down data points... instead of what they can do what cant they do, an where do they prefer to start are a few observations we track.
In the end I use both methods one to validate the other and do look for learning opportunities for the kids. I think there is certainly room in app development to track things that aren't tracked in the apps I have tested. Many seem to track too much and at the same time not enough IMO. Cutting down the noise is good too and focusing a a few key areas seems to do fine. I do love the stats many post here kudos to them.
|
I think the original question should be redirected. (That's often the role of a mentor addressing a student.) My point is on the "best way to scout" is to keep in mind the overall mission of FIRST which is to focus on the educational aspect. So in looking at scouting, we should look beyond competition but to the overall educational mission of FRC. Whether a scouting system is truly successful is really secondary to what students learn in the process of developing and implementing the scouting system. That's why we provide white papers on developing the systems, not a turnkey version of the system itself. We want other teams to build their own systems.
We've found that the students and younger mentors learn how to better mesh data analysis and strategic observation. The draft lists we generate are only a starting point--it helps separate the very best, the middle and the low end, and then we have to rerank the middle in particular with our own observations. It's an education process in the limits of strict data analysis.
The fact is that this is really on the only venue in high school where students can directly learn about how to use the convergence of data and direct observation. Most social science classes convey the observational part through histories and theories, and the stats classes teach the mechanics of data analysis. I don't think a team should miss the opportunity of providing yet another project-based learning experience that brings together several different educational threads. This is what using "Big Data" is about. Decision science and analysis are growing majors in college and FRC grads can be well positioned to perform well in those programs.
So I wouldn't want to leave scouting solely to an exercise in observational judgement. There are other many other venues for that type of experience, but FRC provides a truly unique opportunity that no team should pass up.