tubechallenge.com

hwolge wrote:

tpfkar wrote:
However we are also planning a warm-up event on the evening of Thursday May 17th. This will be along the lines of the idea which seemed to be well-received when posted earlier; a starting station is drawn in advance, and then all the other stations which have to be visited are drawn at the start of the challenge, and you have to plan your route as you go. I don't want that to take all night, so will work out a sensible number of stations. We'll work out which zones to use nearer the time.


May I humbly suggest the following setup for the warm-up challenge:

* Use a fixed start station (and quite possibly a fixed end station as well?)
* Limit it to Z1 & Z2.
* Draw 15 (?) random stations. Enough to make planning interesting.
* Put a time limit on the event, i.e. get the highest number of target stations before XX o'clock. If two teams share maximum number (quite possibly all of them) the quickest time for this number of stations is the winner!

I've started a new thread on this as it warrants its own discussion. Myself, David and Kevin have already been discussing the format, however my gut reaction is that I think these are excellent suggestions - thank you for posting them. We were planning to do the first 3 suggestions anyway, however a fixed finishing time may be a good plan as we are worried about it taking a long time, particularly given people (including myself) are planning on coming down after work.

My current plan is to write a computer program to generate the 15 stations, but do it in such a way that I see the starting station only before printing a batch and sealing in an envelope. Announce the start station. Then start the clock when the envelope is opened at the station, coincidentally just as a train is coming into the platform. This will save faffing (I can't really be bothered writing out all station names onto small pieces of paper), people having to write the stations down, and also be faster - as we are likely to be pushed for time.

I'd like to open this up for comments however. One thing I've been saying on the emails going round about this is that this is a brand new challenge so there is no precedent - this is the time for innovative ideas before the more traditional (perhaps ) challenge on the Friday.

To make it even simpler I'd suggest you (organisers) just decide on the starting station, since this has to be known in advance anyway, and why not the same place as usual? This would also give an interesting choice of starting line/direction (unfortunately invalidating your nice idea of opening the envelope at the moment a train arrives...).

I'd also say it would be nice to give the printing and enveloping of the route to a trusted umpire/referee, if available...

Adding a fixed end station, announced in advance, near an intended gathering point (probably just any predecided pub / McD would do), would make for an interesting finish as well!

What rough time plan are you looking at?

This challenge sounds like more fun every time I think of it!

PS Return tickets from Sweden was around £70 per person...

I think we are looking at starting time of around 7pm, will try for earlier if possible....

I would suggest a finishing time of around 1030hrs, as we have done a number of evening Z1's which take around 3 1/2 hours and this always allows enough time for a post challenge chat and bite to eat

I like the idea of 15 stations and think we could possibly even extend to Z3
You could split the number of stations over zones, say 8 in Z1, 4 in Z2 and 3 in Z3...
Then a points system could operate say 5 points for Z1 visit, 10 points for Z2 and 20 points for Z3... And 50 point bonus for visiting all 15 within the time allocated...

Not sure about the idea of a pre-nominated finishing station....and we need to ensure that it is a challenging challenge
We don't want to have 15 easy stations and whizz around them all in 45 minutes

Looking forward to seeing you all again on 17th / 18th May

I don't think 15 stations would be too little. Extending to Z3 would probably only mean longer time, but not harder planning (lots of double backs likely).

A points system would be interesting. However, it would probably change the challenge into something completely different than initially intended.

I'll try to do some simulations of approximate event time based on number of stations and so on... I'll post any results here!

All good stuff, if we keep bouncing the ideas around I am sure we will come up with an interesting challenge by 17th May

The idea of stations having different values is very similar to orienteering 'score' events which I have taken part in; every station has a value, and you visit as many as you can - with severe time penalties if you are late to the finish. All stations could be in play under this idea - would it be worth going to Amersham and knocking off all the high value stations, or covering all of Z1/2 in the time? Would be an interesting strategy challenge, although different to what I had in mind so far.

The points system would certainly lend itself to planning in advance and I have a gut feeling several teams will come up with similar ideas.

Having random points assigned at the start point (in a sealed envelope) would certainly be an interesting challenge. I do however think we're then moving from "a challenge" to "a competition" even though it's a thin line here...

Personally I prefer a more straight forward challenge like the one indicated by Ric and myself earlier - but I have an open mind!

KISS! (Keep It Simple, St***d)

Can I just chip in and say I have no idea what you're all talking about, and that I am thoroughly confused and bemused as to why you're doing this the day before the Zone 1 Challenge?

Carry on ...

tubeguru wrote:Can I just chip in and say I have no idea what you're all talking about, and that I am thoroughly confused and bemused as to why you're doing this the day before the Zone 1 Challenge?

Carry on ...

I guess its like a little starter just before the main course.
It seems a good idea if several people have travelled and are spending the night in or around London to do a small event beforehand and something a little bit different, just to get the juices flowing for the official Z1C

Soup Dragon wrote:

tubeguru wrote:Can I just chip in and say I have no idea what you're all talking about, and that I am thoroughly confused and bemused as to why you're doing this the day before the Zone 1 Challenge?

Carry on ...

I guess its like a little starter just before the main course.
It seems a good idea if several people have travelled and are spending the night in or around London to do a small event beforehand and something a little bit different, just to get the juices flowing for the official Z1C

Yes exactly that was the thinking behind it and also a chance to warm up and get the the legs moving prior to the main event

hwolge wrote:KISS! (Keep It Simple, St***d)

Can I ask why the term 'stupid' requires asterisks?

Cos it's rood, innit?

Xegactly!

hwolge wrote:I don't think 15 stations would be too little. Extending to Z3 would probably only mean longer time, but not harder planning (lots of double backs likely).

Not necessarily, as depending on the Z3 stations that are randomly generated cunning use of national rail or bus connections could prove logistically better than doubling back.....

Hi!

I've been playing with my computer. I looked at the following setups:

* Fixed start at Oxford Circus.
* 14-15 additional random stations
* Stations drawn from Zones 1, 2 and/or 3

The following rough approximations of a "good route" for each case would be:

Z1+2/14 stations: mean=2h48m, stdev=23m
Z1+2/15 stations: mean=2h56m, stdev=26m
Z2 only/14 stations: mean=3h25m, stdev=22m
Z2 only/15 stations: mean=3h34m, stdev=22m
Z1+2+3/14 stations: mean=3h36m, stdev=35m
Z1+2+3/15 stations: mean=3h46m, stdev=33m

Rule of thumb says that 2/3 of the cases are within 1 stdev from mean, i.e. 5 times out of 6 the result would be less than mean + stdev...

I looked a bit further on the Z1+2/14 stations scenario. It seems that in 95% of the random cases, it'd reasonable for a "good route" to be around 3h30m...

Any questions?

EDIT:
Full simulation results here:
http://lund.dyndns.org/zone1/Simulation.html