Suggestions for the 2013 season

cgipryan wrote:

"3)at the point at which the competition is turned on the rider values will be calculated. They'll be calculated based on a rider's performance in the year preceding that date; ie all races up to one year in the past will be considered

4) the scores of each rider will be totted up to generate averages for each type of stage; they'll be weighted according to the importance of the race, and how many of each kind of stage features in the upcoming race

5) based on that, each rider will be given a value, and the highest-ranked riders will be star riders. so a rider won't necessarily stay a star if they do badly over a season, and the reverse is also true."

It seems to me like you're gonna do much of the thinking for us... I'm pretty worried that this might lead to winning by pure luck...

really? because all we're doing is using the available data to set rider values, like we did last year. the only difference being this time we're using *actual* data, rather than data in our heads.

Gkam84 wrote:

It would be 1/198 if you could only pick one rider, But you get to pick 9, So its got to be divided by 9

Not really... If I had to guess, I'd say the odds are less than 1/22, because the way you're looking at it, you could only choose 22 different teams of 9 - to cover all 198 riders... One team would have bibs 1-9, another team another nine riders, but in reality there are more than 22 different combinations, many more, so I figure the odds are less than 1/22... I may be wrong, but I wish we had a statistician on the forum...

Dave Atkinson wrote:

... all we're doing is using the available data to set rider values, like we did last year. the only difference being this time we're using *actual* data, rather than data in our heads.

I'm wondering what criteria you'll use to determine if a rider is a Star. How many points they scored in a competititon? If they were Top 10 in GC or Top 10 in a Classics race? How many points they scored in the previous race they rode?... Take Westra, for example, Got 3nd in GC at Paris-Nice 2012, had 3rd highest point total in the Roadcc competition, I think, what would you do with him for Paris-Nice next year?...

I still think the amount of Star Riders per competition should be cut down and that the Star Rider criteria, having 4 of them, should be set as an upper limit, not as a requirement, of filling 4 slots, only with Stars...

enrique wrote:

Dave Atkinson wrote:

... all we're doing is using the available data to set rider values, like we did last year. the only difference being this time we're using *actual* data, rather than data in our heads.

I'm wondering what criteria you'll use to determine if a rider is a Star. How many points they scored in a competititon? If they were Top 10 in GC or Top 10 in a Classics race? How many points they scored in the previous race they rode?... Take Westra, for example, Got 3nd in GC at Paris-Nice 2012, had 3rd highest point total in the Roadcc competition, I think, what would you do with him for Paris-Nice next year?...

I still think the amount of Star Riders per competition should be cut down and that the Star Rider criteria, having 4 of them, should be set as an upper limit, not as a requirement, of filling 4 slots, only with Stars...

Dave already explained earlier how the value system for riders works.

My dad and probably numerous others have said "if it aint broken, why try and fix it".

This can be said of the game at the mo. Everyone has ideas but until something within the game blatantly doesnt work why change it. Its easy for newcomers to get to grips with and the more you pour into the game the more complicated it gets thus ruling out a probable large majority of users. You just have to look at the scores in the leagues, teams from pages 24 to 28 have zero points (thats a 1000 users) and if you make the game more complicated these numbers are just going to increase.

stumps wrote:

...Dave already explained earlier how the value system for riders works...

Yeah, I guess I'm looking for more clarity on, not the value of a rider, but whether he becomes a Star Rider or a DS... and whether the 4 Stars per team requirement is up for discussion or whether it'll pretty much stay the same...

Gkam84 wrote:

... Before you break it down, the chances of picking a straight out winner would be 1 in 22...

I wonder if the statistics is right in this... Could it be right to think that the odds would be 1/198 + 1/197 + 1/196 + 1/195 + 1/194 + 1/193 + 1/192 + 1/191 + 1/190, more or less 4.7%, slightly less than 1/20, so very close to your 1/22... I have nothing else to do, obviously...

Dave Atkinson wrote:

cgipryan wrote:

"3)at the point at which the competition is turned on the rider values will be calculated. They'll be calculated based on a rider's performance in the year preceding that date; ie all races up to one year in the past will be considered

4) the scores of each rider will be totted up to generate averages for each type of stage; they'll be weighted according to the importance of the race, and how many of each kind of stage features in the upcoming race

5) based on that, each rider will be given a value, and the highest-ranked riders will be star riders. so a rider won't necessarily stay a star if they do badly over a season, and the reverse is also true."

It seems to me like you're gonna do much of the thinking for us... I'm pretty worried that this might lead to winning by pure luck...

really? because all we're doing is using the available data to set rider values, like we did last year. the only difference being this time we're using *actual* data, rather than data in our heads.

I don't think this is all you're doing, because you are also anticipating how the race is going to go... especially when you say, I quote, "3) It'll mean that rider values will be weighted towards the riders that will do well on certain types of terrain; if a race is all flat stages then sprinters will be more expensive, and so on".
This is where you're practically gonna organize the riders (weighting in their current form and abilities on the specific terrain) according to price, and we're gonna be left pretty much with randomly picking a set of 9 from the pre-organized (according to winning chances) rider list. It sounds to me like you're willing to organize the rider list for a race like the betting odds lists before a race. And that already indicates the riders who have the largest chances of winning and also forces the participants in the game to take chances (lucky ones, basically) if they want to win much more than now, I think...

Dave Atkinson wrote:

enrique wrote:

the reason we all have similar teams at the end of a Grand Tour is that those riders are getting Finish Line Points, GC points and KM points and sometimes PC points...

the reason we all have similar teams at the end of a Grand Tour is that we pick the set of riders who are going to score most highly within the constraints of the rules. changing the rules might change which riders you pick but it won't change the fact that teams converge. it's inevitable.

Another one of the skills is knowing when to dump guaranteed GC/KOM/Sprint/Young points for possible break points.For example in the latter part of the Tour De France I believe Voeckler and Kessiakoff were up there in terms of popularity with Wiggo,Froome and Nibali?

cgipryan wrote:

I don't think this is all you're doing, because you are also anticipating how the race is going to go... especially when you say, I quote, "3) It'll mean that rider values will be weighted towards the riders that will do well on certain types of terrain; if a race is all flat stages then sprinters will be more expensive, and so on".
This is where you're practically gonna organize the riders (weighting in their current form and abilities on the specific terrain) according to price, and we're gonna be left pretty much with randomly picking a set of 9 from the pre-organized (according to winning chances) rider list. It sounds to me like you're willing to organize the rider list for a race like the betting odds lists before a race. And that already indicates the riders who have the largest chances of winning and also forces the participants in the game to take chances (lucky ones, basically) if they want to win much more than now, I think...

So you think that by organising the riders values by how well they did LAST season and depending on the type of race, Small tour, Classic, Grand tour. Then Dave and the people behind the game are some how indicating who has a chance of winning the race??

You know when riders line up, they all want to win and there are no guaranteed winners in cycling. So by setting the values depending on their performance from the year previous means nothing. Just look at Gilbert this season. He couldn't do what he did in 2011

cgipryan wrote:

I don't think this is all you're doing, because you are also anticipating how the race is going to go... especially when you say, I quote, "3) It'll mean that rider values will be weighted towards the riders that will do well on certain types of terrain; if a race is all flat stages then sprinters will be more expensive, and so on".
This is where you're practically gonna organize the riders (weighting in their current form and abilities on the specific terrain) according to price, and we're gonna be left pretty much with randomly picking a set of 9 from the pre-organized (according to winning chances) rider list. It sounds to me like you're willing to organize the rider list for a race like the betting odds lists before a race. And that already indicates the riders who have the largest chances of winning and also forces the participants in the game to take chances (lucky ones, basically) if they want to win much more than now, I think...

you go with picking your riders randomly and i'll go with attempting to pick a squad i think will score well, and we'll see how that pans out.

one thing i've learned from three years running the game is that some people are really good at it. the rules might change a bit from year to year but it definitely takes skill to win, and that'll be the case next year too.

Dave Atkinson wrote:

one thing i've learned from three years running the game is that some people are really good at it. the rules might change a bit from year to year but it definitely takes skill to win, and that'll be the case next year too.

Something I have learnt from playing for 2.5 years. Skill = lots of time to researching the riders and race route, understanding the rules and scoring system, planning well ahead and making the right decisions.

Personally I do not see the need for "Oh crap, picked the wrong DSs for a GT, I need a magic bullet to fix it and be competitive"

This was my starting line up for the TDF
Peter Sagan
Fabian Cancellara
Edvald Boasson Hagen
Samuel Dumoulin
Daryl Impey
Julien Simon
Jonathan Cantwell
Gustav Erik Larsson
Kris Boeckmans

I made some bad transfers during week 1, and was outside the top 100. I improved after that, finishing 3rd, I did not win the TdF comp because I picked Pierre Rolland over Thomas Voeckler on Stage 16.

I do not want to see a change to only bank transfers for one day or lose them. It has caused me problems in the past and will in the future, but it makes you study the race route and plan ahead. Otherwise you reward lazy players with stage wins.

I would like to see the artificial 4 star 5 DS rule removed because:
1. It makes the game a little more complex;
2: Gives players far more options in team strategy and make up.
3. There have been times I would of preferred 3 stars and a couple of high value DSs, but could not as I need to spend points on a 4th cheap star to fill out the roster. Other times, I could afford 5 or even 6 stars.

Happy to see the points value of riders change throughout the season, as long as the point cost distribution remains the same.

Personally I would prefer it if all the riders in a race was magically ranked behind the scenes (so we can't complain and the top rider in a race cost 40 (note all these numbers are just plucked out of the air)) , the 2nd ranked rider 39, the 3rd 38 so on, until in the case of the TDF 179th to 188th ranked riders cost 3.2 and the 189th to 198th ranked riders cost 3 or something similar.

So some unknown 180th ranked DS worth 3 points rides stunning Tour of Qatar wins a stage, has a couple of top 10 places and then is ranked 30th in a weak field in the Tour of Oman and costs 15 points. Does not do anything special in the Tour of Oman, next is Tour of Flanders, and the field is stronger and the rider is now ranked 90th in the race and cost 6 points. Disastrous classic campaign, overall ranking falls and by the time of the Giro is ranked 130th in the race and worth 4 points.

Love the idea of a KOM and Points jersey comp for the GT (GC is a bit naff, rewards the safe players, the KOM is for the risk takers and points if awarded in Giro/Vuelta style is for the consistent picker of stage wins, ignoring the safer jersey points and team bonuses etc).

Not a fan of the 2 riders per team limitation, but see it as necessary evil for TTTs.

TERatcliffe26 wrote:

I guess its not so much the rule, its more if you have that rule then you shouldnt have an overlap in values, as imo a rider shouldnt be classed as a star if they are valued at less then a DS as it defeats the object of the split in the first place

that's addressed for 2013; the star riders are the most expensive ones for any given tour.

What I'm saying is that if you organize the rider prices according to their CURRENT form (not last year's form) and especially according to their chances of winning IN THAT PARTICULAR RACE (WITH THAT PARTICULAR PROFILE), we're going to play more against you than against other players. It's like when you bet, where you have two possibilities: a) you bet against the house because you think they've misjudged one rider's odds; b) you play safe by taking one of the the favourites.
So, basically you're going to make the game easier for everyone by indicating the favourites for that particular race, and you're going to force those who think they know better to take much more chances if they want to beat those who just go with the designated (by you) favorites.
The idea is precisely that I DON'T want to pick my team randomly (I was 6th in the overall table this year, so I hope my teams weren't picked randomly...). Of course, this is just my impression, it might not be confirmed when this system is actually tested, with actually implemented prices per rider and with the obligation to pick 9 riders effectively taking part in that race...

nickobec wrote:

... I do not want to see a change to only bank transfers for one day or lose them. It has caused me problems in the past and will in the future, but it makes you study the race route and plan ahead...

I do want to see a change. I like the flexiblity of using them when you want to, after they've been earmed. I do like the way we accrue them, I just think once you earn them, they should be used at will...

nickobec wrote:

... Not a fan of the 2 riders per team limitation, but see it as necessary evil for TTTs.

I'd be ok with removing this rule. I mean, you have to deal with the consequences of having them in your roster later... Why not go up to 3 per team?...

TERatcliffe26 wrote:

... regarding the 4 star/5 DS split... I guess its not so much the rule, its more if you have that rule then you shouldnt have an overlap in values...

I do think it's the rule... I think there comes a point in every Grand Tour when your Stars won't be the guys scoring the most points and to have to choose 4 Stars hampers you and may prevent you from catching up to the competition leaders, who may be playing it safe at that point in the Tour...

Sometimes you have to resort to picking someone who will provide you with the certainty of GC points, for example, because you have to have 4 Stars... but as TER says above, you'd have gladly put in a DS...

I wish, again, that if it's going to be a requirement to have Stars, that the requirement be brought down to 3 Stars or even 2, but I still prefer the flexibility of having the Star restriction be an upper limit and not a requirement...

Dave Atkinson wrote:

... that's addressed for 2013; the star riders are the most expensive ones for any given tour.

Dave, what are your thoughts on the number of Stars for next year? Are you inclined to keep the 4 Star requirement? Can you drop it down to 3 Star Riders, if they're going to be required to be filled? ...

cgipryan wrote:

What I'm saying is that if you organize the rider prices according to their CURRENT form (not last year's form) and especially according to their chances of winning IN THAT PARTICULAR RACE (WITH THAT PARTICULAR PROFILE), we're going to play more against you than against other players. It's like when you bet, where you have two possibilities: a) you bet against the house because you think they've misjudged one rider's odds; b) you play safe by taking one of the the favourites.

So, basically you're going to make the game easier for everyone by indicating the favourites for that particular race...

Yeah, I see your point... Anyone can have a pretty decent team if they choose the higher valued riders...

Then again, you still have to choose your DS's...

You know, but if that is the case, then I feel more strongly towards limiting the Stars to just 2 or 3... 'Cause it forces, or might force, much more diferent teams than if we have to pick 4 Stars...

cgipryan wrote:

What I'm saying is that if you organize the rider prices according to their CURRENT form (not last year's form) and especially according to their chances of winning IN THAT PARTICULAR RACE (WITH THAT PARTICULAR PROFILE), we're going to play more against you than against other players. It's like when you bet, where you have two possibilities: a) you bet against the house because you think they've misjudged one rider's odds; b) you play safe by taking one of the the favourites.
So, basically you're going to make the game easier for everyone by indicating the favourites for that particular race, and you're going to force those who think they know better to take much more chances if they want to beat those who just go with the designated (by you) favorites.
The idea is precisely that I DON'T want to pick my team randomly (I was 6th in the overall table this year, so I hope my teams weren't picked randomly...). Of course, this is just my impression, it might not be confirmed when this system is actually tested, with actually implemented prices per rider and with the obligation to pick 9 riders effectively taking part in that race...

i don't think you're quite understanding how it would work. for a start the values will be based on a year of form to the date of the race, not just last year's race. And although the prices will be weighted according to what kind of stages are in the race, neither that nor the year's form will be a perfect indicator of race form. there'll still be bargains, and expensive flops.

Also, i'm not sure why you think that the riders most likely to win shouldn't be the most expensive? yes you could do okay just picking favourites, but it's not going to get you the win, and anyway you won't be able to just pick all the favourites because you'll run out of cash. the four most expensive stars for a race would likely use up all your budget.

enrique wrote:

Dave, what are your thoughts on the number of Stars for next year? Are you inclined to keep the 4 Star requirement? Can you drop it down to 3 Star Riders, if they're going to be required to be filled? ...

i'm inclined to drop the split altogether and let everyone pick who they like. the new rider values system will give a more even spread of values so i think more flexibility in who you pick will be necessary. also looking at upping the riders from a single team from 2 to 3

enrique wrote:

1. How about total banking of any accrued transfers? You can earn them 2 a day but use them - after you have them in your account - any day you want to until the end of the compeititon?...

Dave, what are your thoughts on transfers and the penalty fees?...

Dave Atkinson wrote:

... here's how it'll basically work:

1) there's a maximum rider value of 40 and a minimum rider value of 3...

Dave, I love riders that can be gotten for less than 3 points, just so you can have some flexibility to stack your team a little, and so we don't have people resorting to picking riders at 2.3, 2.4 or 2.5 points that aren't riding, and are grayed out... How about giving us a few of these in each race? Surely there'll be no-name domestiques we can afford to do this with... Huh?

enrique wrote:

Dave, I love riders that can be gotten for less than 3 points, just so you can have some flexibility to stack your team a little, and so we don't have people resorting to picking riders at 2.3, 2.4 or 2.5 points that aren't riding, and are grayed out... How about giving us a few of these in each race? Surely there'll be no-name domestiques we can afford to do this with... Huh?

not sure i understand. the point of the 3 credit minimum is that there'll be no point picking greyed-out riders as there won't be any cheaper riders not participating. it could easily be 2.5, or 2, or whatever