New ctc site

[JonT]

Jon, not sure why, but your CTC monitor isn't picking up that Ian is in boat 1 for some reason.

It’s because he appears as Ian Bee in the rower list and Bee, Ian in the boat. No idea why, but it means our tool doesn’t place him in the boat. All a bit tedious to be honest. I’m hoping it gets fixed at their end.

Jon


Sent from my iPhone using Tapatalk

[Ian Bee]

I'd noticed that..... I'll see if it's in my settings to fiddle.

[JonT]

I'd noticed that..... I'll see if it's in my settings to fiddle.

I suspect it is at the CTC end Ian. But as far as I can see you are the only person in the CTC that this is happening to, and I don’t think it happened last month, so probably not in your settings unless you have changed anything.

Oddly, if you look at the results on CTC for last month the names all appear as first name followed by last name. Now they are all switched around, except for yours!

I can always manually edit the data to fix this when I do an upload to our tool (remember this is not automated any more), but obviously that's not a great solution.

Jon

[obirobsan]

It's getting there, now we have the non floating boats listed, just no boat numbers or average time/distance yet.

[JonT]

It's getting there, now we have the non floating boats listed, just no boat numbers or average time/distance yet.

Which of course means the calculations for the new tool will all be thrown out again. Happy days! Having said that, it’s probably as simple as removing the factor that I added in last month.

I’m away at the moment but will take a look when I am back next week.

Jon


Sent from my iPhone using Tapatalk

[Iain]

Jon, will be a little awkward to go back to the old model as you will need to renumber the boats! "Interesting feature" that now the fastest Lwt is in the Lwt seat, while on the old site it was the slowest!

[obirobsan]

Yeah I saw yesterday we had a boat with only a lightweight in it followed by one with a bunch of heavies and a slower lightweight. The new site still is not quite working like the old … I was thinking “that’s the downside of moving off the 20 year old code that had all the bugs worked out”


Sent from my iPhone using Tapatalk

[JonT]

As expected the new tool can't handle the new way that ashore boats are shown on the new site.

I'll get it fixed by the end of the month if I get a chance.

Jon

[JonT]

I restarted my efforts to get our tool running now that the CTC folks have reintroduced boats that are not full. But, unless I have misunderstood what they are trying to do, that feature does not seem to be working properly. For example, Tako has an entry, but is not showing in a boat that is afloat or incomplete. In fact we have no incomplete boats showing. Until they fix this, we don;t have a working tool

EDIT - CTC have confirmed the site is broken.

[JonT]

I have rebuilt the tool (yet again) to deal with the fact that partial boats are now included in the total number of boats. The layout is different and so the code needed to change, but the scoring is now back to the way it was at the start of the year. The tool is under the Utilities menu, or here if you prefer.

[fkoene]

Again: thanks thanks thanks, Jon

[JonT]

The new C2CTC site continues to be the gift that keeps on giving.

You may remember that in May the site changed so that incomplete boats were reintroduced into the leader board. This meant that I had to undo all of the changes that I had made to compensate for the incomplete boats being missing. The site has changed yet again and not the incomplete boats are shown for the current event but are deleted when the event is complete. This changes all of the calculations depending on whether you are on the last day of the competition or the first day of the following month. Trying to deal with this by showing the scores during the month and then the same scores immediately afterwards would be incredibly complex (verging on impossible). Remember also that the update to our tool is now a manual one because the site requires you to be logged in to access the scores and so can't be done remotely by our website. If we stuck with te tool as it stands (calculating based on incomplete boats) that would mean I would have to be certain to login to the site and grab the data on the last day of every month. I can't guarantee I will be able to do that.

I'm open to suggestions folks. If we view the tool as a way to make Fred's life easier, then the solution is just to grab the data after the month is complete and post it for Fred. If we view the tool as a good way to track how we are doing, then the calcs need to be based on incomplete boats and rely on a snapshot being taken at our end right at the end of every month.

It's great that someone has taken over the site and is managing it, but they are not making life easy at the moment. Of course, they don't see our tool as a 'use case' for the website, so I don't really have any cause to grumble.

Jon

[Iain]

I would vote for whichever is easier which I suspect is ignoring the incomplete boats from the site, but if it is possible, can the "results" be shown so that people's relative position can be determined and potential intra-team gains assessed?

[stumpy]

Jon what ever you decide to do is just fine by me, as always a huge thank you for your wonderful work in keeping the site running.
Ps I’ve managed to log in this morning to the CTC site with no problem after yesterday’s failure.

[fkoene]

I don't mind doing some more calculations, but when, like you said Jon, incomplete boats are deleted when the month finishes I somehow need to know who finished the challenge.

Verstuurd vanaf mijn FP3 met Tapatalk

[JonT]

Maybe the way to do this is to abandon the boats as displayed in the CTC and instead work out who is in a floated boat and who is ashore just from the list of rowers. That way we always have our own version of "the truth". That would mean that a score would always need to be based on the number of floated boats in the CTC rather than the total number of boats and would also mean that every ashore boat gets the same score (I think). Let me have a ponder.

Jon

[JonT]

I will try to write a new version of the tool next week which works no matter how the website deals with boats that have not floated.

Iain - I’m counting on check the logic for me


Sent from my iPhone using Tapatalk

[JonT]

I have updated our analysis tool (again!) hopefully to have the tool working irrespective of whether the CTC site lists boats which are afloat or not. We have to do this because up to the end of the event boats which have not floated are listed, but they are deleted after the completion of the event. This messed up the last version of the tool which got confused about the number of boats in the event and whether boats were afloat or not. Hopefully this is now fixed. You can see that the table now shows all rowers and whether they are afloat or not. The score is now based on the total number of floating boats in the CTC. Your boat position is subtracted from that to give a score which is then multiplied by a factor (currently 1.4) which we had to introduce because of the large number of The Diamonds boats messing up the old scoring system. Boats which have not floated are allocated 50% of the points of our lowest float boat.

This has taken a couple of days to get right, and I haven't done a massive amount of testing. Please do shout if you see anything wrong.

The new tool is here -> https://www.freespiritsrowing.com/c2ctc/c2ctc_analysis and also under the Utilities menu of our website.

[Iain]

Thanks for your hard work Jon. Seems to be working for the floating boats, but the next 3 people have been selected for boat 5 instead of 3. No impact on points, but looks like somethings gone wrong (although the site gives wierd results for people still on the bank). I don't know how the hardest potential allocation works, is the maximum number of boats prioritised over boat position? This is where a lightweight or lady is needed in a lower boat, but is faster than a heavy currently in a higher boat.

[JonT]

Yes, the CTC site does very odd things with rowers that are afloat. I haven’t checked, but I suspect the non-floating boats are just filled using an odd combination of rower type (H/L/F) and the order that the row was completed.

Maybe it would be clearer for our tool not to show a boat number if the rower is ashore.

As you say, none of this makes a difference in terms of allocation of points because all ashore boats get the same score.

Thanks for taking the time to have a look and give feedback

EDIT - I've fixed this so no boat numbers are shown for boats that are not afloat. The C2CTC site is down at the moment so I can't try this with updated data. It works for yesterday's data though.

[JonT]

After being down for most of the day, the CTC site is back up and running. I have added a line in our tool which shows when the data on their was last updated, given I have to update it manually from time to time through the month.

Jon

[Wolfmiester]

cheers Jon, fabulous work. thank you

[JonT]

A while back. Iain correctly pointed out that, with the new setup of the CTC site and the way we now calculate scores, it was possible that a boat would score a lower score when it floated than it would have done if it had stayed ashore. With boat 4 being in last place this has now happened - being 154th out of 154 gives 0 points. Staying ashore would have given 23 points - half of the points of the the lowest boat afloat. I need to fix this. Next time I'll not only listen to Iain, but also actually do something about it. Hopefully it won't take me too long to get this sorted

Meanwhile, ignore the scores in the table Fred.

Jon

[Iain]

Its ironic that after Christian beat me by 0.1S I am in that boat!

[JonT]

A while back. Iain correctly pointed out that, with the new setup of the CTC site and the way we now calculate scores, it was possible that a boat would score a lower score when it floated than it would have done if it had stayed ashore. With boat 4 being in last place this has now happened - being 154th out of 154 gives 0 points. Staying ashore would have given 23 points - half of the points of the the lowest boat afloat. I need to fix this. Next time I'll not only listen to Iain, but also actually do something about it. Hopefully it won't take me too long to get this sorted

Meanwhile, ignore the scores in the table Fred.

I think I've fixed this. I'm sure I could have come up with a more elegant solution than the one I did in a rush.

Boat 4 now has 23 points, which is half of the boat above (rounded) and the same as the ashore boats. Fingers crossed it keeps working as this month's boats start to build. I think I've got that covered, but time will tell.

NB - I have only applied the fix to the last boat afloat. If I was really clever I would write something that rippled up through all of our boats and worked out which ones would be better ashore - for example, technically boat 3 would have been better taking half of boat 2's points instead of floating. But I can easily imagine months where only the first boat was given its actual score and all other boats got half of boat 1's score. That doesn't seem very logical to me, so I propose leaving it as it is currently (i.e. the lowest scoring boat which is afloat gets the largest of its calculated score based on its position or the 50% of the points of the boat ahead of it).

Shout if you think I've missed something or made an error.

Are we having fun yet?

Jon