Does CD affect alliance selections?

[XaulZan11]

This is something that I've been thinking about for awhile and the other day in my social psychology class we talked about how people's preconceptions affect thier perception of the person. With all the predictions and discussions about the various teams before Atlanta, I thought it would be a good idea to post it now.

First, some social psychology. It has been supported by many studies that the first information you hear or know about an individual or object, carries more weight than later information (also known as the primacy effect). Secondly, people form schemas, or a mental framework or set of expectations we expect to see, once we know something/anything about an individual or object. Finally, there is the confirmation bias that is the tendency for people to look for examples that support their beliefs. For example, if you think someone is nice, you pay more attention to the incidents where they are nice and ignore the times they are mean. There was a very interesting study that took a class and told them they were going to have a speaker come talk to them. Half the class were told the speaker is a famous expert on the subject, speaking at plenty of universties and conventions. The other half was told that the speaker was a graduate student that was practicing giving speeches. Although they listened to the same talk at the same time, the two groups had significantly different opinions on the speech. Those who thought he was a famous expert on the subject, rated the speech significantly better than those who thought he was just a graduate student.

So, back to robot. I'm think these factors may contribute to people's perception of the various robots and thus could affect alliance selections. Suppose you read a thread about a team that says that they are simply amazing and have the best launcher. When scouting at the competition, you already have a schema set in place (that they have great and have a great launcher) and the confirmation bias says that you will look for examples that fulfill that schema, so when they do a successful launch, you think 'I knew they were good', but when there launch hits the overpass you think, 'well, that team was in thier way, or that was just unlucky'. On friday night you go through the scouting data and see that this team, suprisingly, isn't at the top of the list. Instead of thinking that they may not be as good as you once though, you just remember the good launches and the good matches. Come alliance selections, you may be more willing to select that team because you already had a schema set in place (from CD) and you looked for things that fulfill that schema because of the confirmation bias.

To use the study on the speaker as example. The information the two groups were told (the speaker is a expert vs just a grad student) is like the information here on chiefdelphi (the prediction thread is a great spot to see people talk about how great teams are). The speech can be compared to the team's matches and the rating of the speech is the team's scouting.

Its just a theory that I've been thinking about. I don't think that we shouldn't discuss how good teams are (that wouldn't be any fun!), but I think it is something that people should keep in mind when determing how good a team is.

[Stephen Kowski]

....yes and it also works like that if you just happen to like a team....or have been partnered with them in prior matches/competitions

A good strategy team should be able to weed out the hype from what a team actually can do....also they should be able to find those teams that are flying under the radar (one of my favorite things)

on the whole I believe though that this is correct, that it influences things, but I do not think it is the definitive reason someone is picked

I've never heard in a strategy meeting, well someone from CD said they were good so I think what you saw today was an aberration and you should disregard it

[Laaba 80]

I completely agree with this. It goes a little deeper though. We had an incident this year where our drive crew had a bad experience with a team that was rather high on our list. After that we really did not want to work with them, even if they had the best robot. I agree with your thinking, however it can involve other situations like I mentioned.
Joey

[smurfgirl]

I agree with all the stuff you said about psychology- it is true, and I see it in myself and others. With that said, however, I don't think alliance selections are quite the same. For example, I went into the CT regional with a predisposition towards working with team 230, since we had just won the scrimmage with them, and our teams get along well. After coaching and scouting galore, the head scout on our team and I agreed that we would play better with 175, and it was a smart move not to put ourselves on a different alliance than our biggest opponents at the regional. Ultimately, it is robot performance etc. that makes teams decide on alliance partners- what you see at a given competition is what affects your choices.

[XaulZan11]

Originally Posted by smurfgirl

Ultimately, it is robot performance etc. that makes teams decide on alliance partners- what you see at a given competition is what affects your choices.

I agree that teams pick based on the other teams performance at the competition. My point is saying that how we percieve a team, or about thier performace, is affected by our preconceptions (what is said here, history of team...). So, unless a team makes selections based on only objective data, these preconceptions may play a major role.

Also, these are not concious things that people are always thinking about and aware of. In the study of the speaker, the listeners did not conciously think, 'well, I'm not sure if this is good, but I'll give him the benifit of the doubt and say this expert is good'. They asked the subjects to guess how tall the speaker was and those who were told he was an expert guessed that he was 2 inches taller than the group that thought he was just a grad student. I doubt they conciously thought that. This is why no one says, 'well, I heard on CD that their good so thier preformance today was just a fluke'.

[David Brinza]

I think the "smoke and mirrors" from CD posts clears up by Friday afternoon.

Scouts try to get some insight into teams from across the country from CD. There are other tools (like the rankings and stats from Karthik and SamSofa330) that help scouts develop a better picture of the teams' capabilities. The fact that those tools are posted on CD makes browsing these threads even more worthwhile.

[AndyB]

I DO think that CD affects team's images and I do think that it helps in most cases.

For instance, I never really looked at 1646's robot, I didn't really know anyone on the team, and I had never really played a match with them. But when I saw they were in our division this year, they struck me as a team I should match, solely because Kyle Love is with them and he is an active CD member...

There are MANY other circumstances like this that I have run across, this is one that came to mind.

Of course, I try not to allow it to affect the list I make for alliance selections, but I'm sure it does happen a little.

[Danny McC]

From CD I tell the people on my team which bots to pay most attention to. But I think that solely picking someone because of what they say they can do on CD or because you are buddy-buddy with them on CD may hurt you in the end. Also, if when we get there they see bots doing well we look to pick them. But about your question, I believe that CD does have SOME effect on the pickings.

[Akash Rastogi]

Originally Posted by Danny McC

From CD I tell the people on my team which bots to pay most attention to. But I think that solely picking someone because of what they say they can do on CD or because you are buddy-buddy with them on CD may hurt you in the end. Also, if when we get there they see bots doing well we look to pick them. But about your question, I believe that CD does have SOME effect on the pickings.

Unless you are really just in it for the fun of it. If there's a team that truly, and I mean truly, does not care about winning and just wants to have fun with their best-friend-team and picks them then I really don't see the harm in that.

+.02

[ttldomination]

People become more aware about the bots and strategies on Chief Delphi. So it can go both ways.

Teams that have a solid strategies, will only gain information about other robots and that may help them make much better decisions. But the decision isn't affected much.

But if a team changes it's strategy due to something they see on Chief Delphi, then the teams that they pick may change depending on how different the two strategies are.

[fredliu168]

I prefer to refer to this as marketting, because I'm weak at psychology.

CD allows teams to showcase their robot and their teams. When regionals or divisions come around, the teams well known on CD will probably be the first to get scouted. During alliance selections, the representative, if given a scouting sheet with no statistics, is more likely to go with a team he or she knows.

Therefore yes, I do think it affects alliance selections, great point.

[thefro526]

I've done the alliance pickings for my team this year and I will say that some of my decisions are influenced by the people on the teams. Like in NJ we picked 708 because the guys on their team were rally cool and would do whatever we needed them too. I've never been in a situation, up until now, where I've actually known people from CD and competed with their team. I know at the Championship this may come into play where a team who I know from CD and another team will be ranked similarly in terms of statistics. If I know people on one team and no one on the other I would be more likely to pick the team who I know because in theory we would work better together but the opposite is also true. If I know you from CD and you're jerky or whatever (very few people like that on here) and I would find it hard to work with you then I would pick the team who I don't know.

[galewind]

I see Chief Delphi as the "Wikipedia" of bot info -- it's a great place to get some surface knowledge, but you need to dig deeper to get the real nitty-gritty. Thankfully, couple CD with The Blue Alliance and SOAP108, and you've got a great opportunity to back up the talk with some actual research. Watch teams compete, check out their stats -- who they were with and against, etc.

What it all boils down to, however, is that all that info should only be used in an initial glance -- you have to see how the teams stack up against the teams currently at the event, and I don't mean by rankings. There are always a few teams that slip under the radar that perform exceptionally well but don't pull out the best records because they were consistently against powerhouses.

[Wayne C.]

CD is a rumor mill. A few facts, lots of opinions and lots of regional bias. Better to gather your own information by efficient scouting and think for yourself. You will be much more satisfied with the outcome.

WC

[Jherbie53]

I believe that CD has the affect that teams get "legend" status and newer teams might look at teams like 71, 67, 111, and 1114 differently because of it. If you are new and you look on CD and see that 1114 is the only team to win 3 regionals twice, then you are going to look at them as "THE" team to beat. When you are at a regional and the MC says that 71 has won the championship 4 times, you start dreading playing against them. Most of the time the "legend" teams are going to back up the hype put out there about them, but if I was a new team and did hear about 1114, it would definitely make an impact on how I view them.

Most people at the competitions will look at CD and the Blue Alliance for pre-scouting, but most will make their decision about teams based on the way they preform on the field and not from the hype about them.