Freezing Saddles 2015

[cvcalhoun]

So it’s agreed, then? Ten teams, all of which are officially slackers?

[hozn]

@ShawnoftheDread 100336 wrote:

Miles from May and June. Since we are trying to encourage cycling through the winter, why not use many riders’ peak months as the basis for the handicap?

I like this idea: use peak season as the metric.

And, yeah, it will be the slackers team again for me. All (or most) of the social benefits without the competitive part! I like competition (e.g. racing), but when the competition is won by more time in the saddle that means it is competing with things like “time with the kids” (for multiple months) and I know I am not that good at letting go. So, slackers team for me!

[Steve O]

I did some statistical chicanery based on last year’s results. Here’s what I learned, which I think supports my proposal of modified randomizing.

Last year the 9th place team (mine ) ended up with 63% of the points of the first place team (I did not include the Slackers). That was using a handicapping system of some inscrutable kind invented by the cannon.

I ran 8 monte carlos on last year’s individual scores (I added two fictional riders to make 100 exactly; a strong one and a wimpy one) with 10-person teams. The results were (last place score as % of first place; higher % = more balanced):

[TABLE=”width: 64″]
[TR]
[TD=”class: xl64, width: 64, align: right”]41%[/TD]
[/TR]
[TR]
[TD=”class: xl64, align: right”]57%[/TD]
[/TR]
[TR]
[TD=”class: xl64, align: right”]57%[/TD]
[/TR]
[TR]
[TD=”class: xl64, align: right”]58%[/TD]
[/TR]
[TR]
[TD=”class: xl64, align: right”]63%
[/TD]
[/TR]
[TR]
[TD=”class: xl64, align: right”]63%[/TD]
[/TR]
[TR]
[TD=”class: xl64, align: right”]72%[/TD]
[/TR]
[TR]
[TD=”class: xl64, align: right”]75%[/TD]
[/TR]
[/TABLE]
Half of the randomized results were as good or better than last year’s results, but with 10 teams included instead of 9.

I then ran 8 Monte Carlos on the modified randomization. Modification: I took the top 20 riders from last year and assigned them 1&20, 2&19, 3&18, 4&17, etc. I then randomly filled out the teams with 8 other riders.
Here are the results:
[TABLE=”width: 64″]
[TR]
[TD=”class: xl66, width: 64, align: right”]62%[/TD]
[/TR]
[TR]
[TD=”class: xl66, align: right”]67%
[/TD]
[/TR]
[TR]
[TD=”class: xl66, align: right”]70%[/TD]
[/TR]
[TR]
[TD=”class: xl66, align: right”]72%
[/TD]
[/TR]
[TR]
[TD=”class: xl66, align: right”]75%[/TD]
[/TR]
[TR]
[TD=”class: xl66, align: right”]76%[/TD]
[/TR]
[TR]
[TD=”class: xl66, align: right”]78%[/TD]
[/TR]
[TR]
[TD=”class: xl66, align: right”]83%[/TD]
[/TR]
[/TABLE]

Only one out of 8 was less competitive than last year. Or zero depending on how you account for 10 vs. 9 teams.

There you have it. There’s a 50/50 chance that random selection of teams last year would have been at least as competitive as the actual teams were. A modified randomization scheme would have performed better in virtually all cases.

I propose modified random. It eliminates potential sandbagging and also eliminates the need for newbies to estimate or people to worry about changing circumstances, etc. It’s easier to implement and is highly likely to result in teams that are at least as competitive as any “system” that we try to devise.

[Arlingtonrider]

@elwbikes 100294 wrote:

I volunteer!!! This is what I get paid to do, so I volunteer.

Fantastic! Thank you so much!! That will be awesome! It’s the main happy hour at the end of the competition that I was most concerned about getting a larger venue for. Would also love to have enough space to accommodate the kiddos a little better at that event if we can manage it. Thanks so much, Emma!

I was thinking of asking about Thursday, Jan 8 or Thursday, Jan 15 at Cap City for the kickoff meet and greet happy hour, if we can get the room. People could meet some of their teammates and other competitors, and the Golden Nugget for most miles ridden on Jan 1 would also be awarded.

Would Jan. 8 or 15 work for most of you? Any preference? A weeknight would allow you to swing by on your way home, and they have all the usual happy hour specials. We’d probably start around 5 pm, but any announcements and the award would be later. (Teams may or may not be announced earlier. Mr. Cannon handles that.)

[Rod Smith]

@hozn 100342 wrote:

I like this idea: use peak season as the metric.

And, yeah, it will be the slackers team again for me. All (or most) of the social benefits without the competitive part! I like competition (e.g. racing), but when the competition is won by more time in the saddle that means it is competing with things like “time with the kids” (for multiple months) and I know I am not that good at letting go. So, slackers team for me!

Understood, but if you were on my team last year you would have been the second highest placed team member. And Karen Smith had more points than you. I think I’ll join the slackers. We can win this thing!

[Anonymous]

@Rod Smith 100348 wrote:

I think I’ll join the slackers. We can win this thing!

NOOOOOO!!! We CAN’T WIN!!! We WON’T TRY!! I have home remodeling projects that have been procrastinated too long, and non-mileage fitness/cycling goals I would like to pay more attention to. UNDER NO CIRCUMSTANCES will I get sucked into just going out and riding some more miles today because my teammates are going above and beyond and i don’t want to let them down, and the people on that other team are also going above and beyond and slowly but surely catching up to us and I have to do my part to not let that happen!

[rcannon100]

@Steve O 100343 wrote:

I propose modified random. It eliminates potential sandbagging and also eliminates the need for newbies to estimate or people to worry about changing circumstances, etc. It’s easier to implement and is highly likely to result in teams that are at least as competitive as any “system” that we try to devise.

Love this.

But…. you are doing this with a known set. You are doing this knowing who the top riders are. Will it blow it up if you dont know who the top riders are. Take, for example, a fictional rider who has never played Freezing Saddles before: Tim Kellie (no relation to our Tim Kelley). If Tim Kellie were to join and be assigned to one of the teams in the modified random way that you suggested, and if Tim Kellie hypothetically was a top cyclists – assigning him to a team would pretty much blow up the game, right?

If that is true, then is the solution simply to still have cyclists self report their average mileage. Use last years top riders in the modified distribution + add in any other top rider who happens to just hypothetically appear. After the top riders are accounted for – then its random?

[Mikey]

@Arlingtonrider 100347 wrote:

Fantastic! Thank you so much!! That will be awesome! It’s the main happy hour at the end of the competition that I was most concerned about getting a larger venue for. Would also love to have enough space to accommodate the kiddos a little better at that event if we can manage it. Thanks so much, Emma!

I was thinking of asking about Thursday, Jan 8 or Thursday, Jan 15 at Cap City for the kickoff meet and greet happy hour, if we can get the room. People could meet some of their teammates and other competitors, and the Golden Nugget for most miles ridden on Jan 1 would also be awarded.

Would Jan. 8 or 15 work for most of you? Any preference? A weeknight would allow you to swing by on your way home, and they have all the usual happy hour specials. We’d probably start around 5 pm, but any announcements and the award would be later. (Teams may or may not be announced earlier. Mr. Cannon handles that.)

I understand CAP City has their big happy hour on Thursdays. Could we do it Friday Jan 2, (closer to the start) if we did it somewhere else? I’d be fine at either, just putting it out there.

[Mikey]

@rcannon100 100357 wrote:

Love this.

But…. you are doing this with a known set. You are doing this knowing who the top riders are. Will it blow it up if you dont know who the top riders are. Take, for example, a fictional rider who has never played Freezing Saddles before: Tim Kellie (no relation to our Tim Kelley). If Tim Kellie were to join and be assigned to one of the teams in the modified random way that you suggested, and if Tim Kellie hypothetically was a top cyclists – assigning him to a team would pretty much blow up the game, right?

If that is true, then is the solution simply to still have cyclists self report their average mileage. Use last years top riders in the modified distribution + add in any other top rider who happens to just hypothetically appear. After the top riders are accounted for – then its random?

Fair does not always equal even. As long as we are under the same rules (quote mileage, etc, be honest) then it is what it is. The more complex your algorithm for determining teams the more disappointed you will be when it doesn’t work.

Or you let the captains do fantasy bike commuting, and draft their teams each wednesday.

[rcannon100]

@Mikey 100359 wrote:

Or you let the captains do fantasy bike commuting, and draft their teams each wednesday.

That would be fun. We CAN do a draft as well. It takes time – that’s the problem. But when I used to play local Ultimate, we had a draft every season. Allows coaches to form their own teams. Have people register with self reported handicaps, and then its up to the captains

This does not scale tho. Last year we grew 100%. If we do that again, the draft could be very long. It is possible to stop-clock a draft (you have 20 seconds to pick – if you dont pick, you simply get the next top draft pick).

[americancyclo]

I like Steve’s idea with an announcement of the teams at happy hour. That would lend a bit of draft excitement to it without people playing favorites.

[Arlingtonrider]

@Mikey 100358 wrote:

I understand CAP City has their big happy hour on Thursdays. Could we do it Friday Jan 2, (closer to the start) if we did it somewhere else? I’d be fine at either, just putting it out there.

I haven’t spoken to the Cap City folks yet, so I can ask for any day that people would like. We just wouldn’t have the great deals on beer and appetizers. So please chime in on preferred dates. If Cap City doesn’t work out, we can try Crystal City, but the last time I checked there, it was very expensive to get a separate room.

[Mikey]

Each of the captains hide color-coded tickets throughout the metro area. Find a ticket you are on a team, and off to visit an excentric candymaker and his slave-run chocolate/murder factory.

[dasgeh]

@Arlingtonrider 100347 wrote:

Fantastic! Thank you so much!! That will be awesome! It’s the main happy hour at the end of the competition that I was most concerned about getting a larger venue for. Would also love to have enough space to accommodate the kiddos a little better at that event if we can manage it. Thanks so much, Emma!

Might be worth asking the Crystal City BID. They can be super helpful and have connections in their community. Plus, Crystal City is a bit more central…

[83b]

@dasgeh 100366 wrote:

Might be worth asking the Crystal City BID. They can be super helpful and have connections in their community. Plus, Crystal City is a bit more central…

The Crystal City Sports Pub has several large rooms that they open to groups. And I recall that the BW3 and Bailey’s also will reserve areas for larger groups.

[Author]

Posts