Ranking Position Numbering Poll

[David Hart]

Dear Forumites,

Over the years, the way that the Ranking has handled verifying pieces, especially those at the top of the tree, has changed several times.

At the moment, the way things are handled is that all times are accepted into the Rankings, irrespective of speed etc. If it is not uploaded direct from a LogCard using the LogCard Utility (LCU), or from RowPro, or from a race, or it does not have a PM3/4 verification code then it is classed as IND – which effectively means unverified.

If your time is unverified AND it appears in the top 3 AND it is fast compared to times from previous years, it currently does not receive a number next to it. So, picking a random example, the Men’s 60-69 Hwt 5,000m Rankings currently looks like this:



The question is: is this system OK, or is it more confusing than just numbering everybody as normal with the implicit acceptance that IND times are non-verified? Is it unfair to people who do not have a PM3 or PM4 and who row at home? The vast majority of pieces are currently IND, although this may change as more people use the PM3/4 verification Code or the LCU. Let us know what you think in the poll, and post any further questions or comments in the thread here.

Best,

David Hart
Concept2

[johnlvs2run]

It is very easy for anyone to get an Ind-V provided they have a pm3 or pm4 monitor.
For example, a few months ago I got an Ind-V for a time of 6:39.1 for a 2k.

I put the results from my 2k on the pm3 in the rankings as a test, then removed it.
My capability was probably about a minute slower than this, but there is the time.



I feel the best way is to include all times in the rankings, as long as the times are
not obviously suspect by being out of the range of what would be physically possible.

A good policy would include honest times regardless of where people have done them.

[michaelb]

I can't tell for sure, but I think John is referring to the bug/cheat that people with model D's and PM3s can set them to think it is a model C and generate false, too fast results, and still get a verified time. So that should be fixed in the firmware/verification process.

I didn't like this change when C2 did it, but it seems to me the problem has actually gotten worse then it was before: more and more times are showing up at the top of the rankings that are unverified (some plausible and some not plausible). So now I think it makes more sense than it did beore, and I wouldn't mind the verification requirement going down to the 5th spot.

It might help to have an error message if someone tries to rank a piece that is clearly out of bounds, say the 4 min 2k times. I assume most of those are typos or are people with the incorrect setting on the PM3.

[Citroen]

The LogCard Utility has all the data needed to submit an entry as IND_V apart from the verification code. So shouldn't C2Log be equivalent to IND_V. One problem with that is for the folks who only have an OmniKey/Athena card reader and no PM3/PM4 at home. They can't generate the verification code without using paper/pen (and sweat proof ink) at the gym.

What I'd like in the LCU is a way to automatically rank the ranking pieces (based on a setting in the utility or a setting in my profile). So I don't have to go searching to see if the one I've just rowed is faster than all the earlier workouts of the same time or distance.

One problem with the whole system is folks logging entries that are faster than the current world records (probably because their PM3/4 is configured for modelC and mounted on a modelD - which gets super quick results). You need some form of checking that the numbers aren't wildly out of bounds. What are the odds of getting a time faster than 5:48 from a normal punter logging a normal 2K? There aren't many Bentons out there. And if there are they won't dislike having to do some extra verification to log their time.

Other than that the system as it is, mostly works as it is.

[johnlvs2run]

I did not use any bug/cheat by changing the model of erg, as Michael is familiar with and suggested.

I simply rowed the 2k distance, got the time from the pm3 monitor, entered the Ind-V and took a photo of my time in the rankings.

It seems to me that people should want to allow all honest times to be included in the rankings, regardless of where people do them; and to not give exclusivity to people just because they have inside knowledge, are computer experts or who are using a particular program.

[David Hart]

The LogCard Utility has all the data needed to submit an entry as IND_V apart from the verification code. So shouldn't C2Log be equivalent to IND_V.

Just to confirm that this is indeed the case at the moment: if it's C2Log it's classed as a confirmed piece, so would always receive a place number.

[TabbRows]

Speaking as someone who's in the middle of the peleton, so to speak, and doubts will ever make the top 10 much less the top 3, I think those with times that do appear in the top 3 ( and maybe in the top 20 based on number of ergers out there) need to have some form of verification for their times. Not that it effects me at all, but when I look to see just how fast others in my age group are rowing, I want to be sure in my mind that those time are legit. If you're over 55 and rowing sub 7 2K's that's fantastic! But if you're over 55 and rowing sub 6:30's you'd better be able to back it up! And I'd think you'd want to do so!

[johnlvs2run]

Tabbrows,

I'm glad you can rest assured my 2k time is legitimate, since it has been verified.

By the way I'm reasonably confident I could lower that to sub 6:30 with some practice.

[Redsculler]

I do understand why someone would want to post an incorrect/false time - the primary person that they are cheating is themselves.

I suppose I am just a naive innocent abroad. If I can get myself back down to sub 8' for a 2k I will be so so happy. Happy to have any times that I post verified if need be. I have a logcard reader and connections and so on - just got to get round to setting them up and connecting them, then I can use the system properly.

When I started rowing the erg at my then club was one of the old bicyle wheel and flaps jobs. I love the new ones. My current erg is a model c with a PM3 upgrade. The monitor is superb.

[Redsculler]

Whoops, that should read 'do not understand'

[Nosmo]

As long as the policy is clear the first three are fine with me. When I first started using the rankings I didn't know why there were some unnumbered times at the top. Changing IND to UNV for unverified and numbering everything would be clearer and simple.

[Sheepster]

I agree with ranking everyone, although I don't really have a strong opinion about this. I actually voted that very fast top three times should be verified. However, I do agree that it is confusing interpreting the rankings as a newbie. Particularly when there are inconsistencies in applying the ranking rules like for example: why does a time of 4min/817 (2:26.7/500m split) not a get a ranking number (lightweight women/30-39) when that time doesn't seem very fast. (The previous year it would have been in 12th place.) Especially given that the top 2000m time (8:03.1 min) for the same age/weight group which has a much faster split is also unverified but is ranked.

Since I started out rowing with model C/pm2 where it was impossible to get a time verified it does seem a little unfair to those people who don't have access to verification to not be ranked. On the other hand, if someone rows a world record time (or very close to it) or tries to post a time that is humanly implausable, it seems reasonable that those times should be verified to get a ranking. But the way it is now - I'm not really sure that the top unranked times are always truly "very fast."

It seems simpler to rank everyone with a "reasonable" time, indicate those who are not verified with an UNV and then have a drop down box in the search that allows people to only have verified times displayed.

[data_duck]

I am many times in the top 3 since I'm in the women's 60-69 HWT category and maybe there aren't a lot of us. Many times I don't know when I'm going to do a fast piece. Sometimes the effort just happens. I row on RowPro for most all of my rows and RowPro is in the comments column on most all of my rows. So far I've gotten a number on all my rankings. I would hate to go to the gym to row and have someone watch me. I would never row as fast in the sweaty gym and would hate to have someone stand around while I might try to do a PB. Most of my PBs happen when rowing with someone else on RowPro in the comfort of my home with fan and airconditioning turned on. I guess I could figure out the IND-V thing to get verified if I had to but I won't be breaking any world records. I am happy to be just somewhere near the top most of the time.

[rlk]

Unfortunately, there are a few rankings posted that don't make much sense -- they're so much faster (or so many meters more) than anything else that it's all but impossible to believe them. For example, for a while someone had listed something like 10,700 meters in 30 minutes, which is such a ridiculous number (well under a 1:30 pace), when there are very few people who can even do 9,000 (1:40 pace) that it just makes the whole ranking system look silly. Remember that power output increases as the cube of the speed -- 1:30 is about 1/3 more power than 1:40 -- and it's just hard to believe that there's someone out there who can produce that much more output than everyone else.

That said, IND_V is hardly foolproof -- someone who really wants to cheat in a particular bracket (as opposed to the "absolute" world records) could always find someone to row for them -- and I don't think we want to get too much into rules for their own sake.

Maybe there needs to be some kind of human review for the top times, and if something really looks fishy, someone just needs to investigate.

[Citroen]

For example, for a while someone had listed something like 10,700 meters in 30 minutes, which is such a ridiculous number (well under a 1:30 pace), when there are very few people who can even do 9,000 (1:40 pace) that it just makes the whole ranking system look silly. Remember that power output increases as the cube of the speed -- 1:30 is about 1/3 more power than 1:40 -- and it's just hard to believe that there's someone out there who can produce that much more output than everyone else.

That's precisely what you get on a model D machine with a PM3/PM4 configured for a model C. (The other way round you get nothing, the monitor won't register a stroke when a PM3/PM4 configured for model D is mounted on a model C.)

Those values are so clearly out of bounds that the ranking system should simply reject them as invalid on input, before the user gets a chance to rank them. It already has some processing for handling a single entry of 100000+.