Upvote Upvoted 129 Downvote Downvoted
1 2 3 4 ⋅⋅ 22
Fantasy TF2: Let’s do this!
posted in Projects
1
#1
0 Frags +

The public beta is now available here:
http://beta.fantasytf2.com/

ESEA Season 19 is the first league and there's only about 17 hours to go before the first match from the time of this post so mosey on over and pick yourself a team.

There are some players not included yet, Champagne and Cocaine are still uncertain as far as I know and hawaii beach vacation needed to confirm players last I heard. I'll make any necessary updates as soon as I know what's going on there. A big thankyou to Alkaline for helping with the roster information.

There will almost certainly be bugs, you can report them here or in the steam group or message me direct and I will get on with fixing them.

The FAQ should answer most of your questions but just ask anything you're still puzzled about.

The public beta is now available here:
http://beta.fantasytf2.com/

ESEA Season 19 is the first league and there's only about 17 hours to go before the first match from the time of this post so mosey on over and pick yourself a team.

There are some players not included yet, Champagne and Cocaine are still uncertain as far as I know and hawaii beach vacation needed to confirm players last I heard. I'll make any necessary updates as soon as I know what's going on there. A big thankyou to Alkaline for helping with the roster information.

There will almost certainly be bugs, you can report them here or in the steam group or message me direct and I will get on with fixing them.

The FAQ should answer most of your questions but just ask anything you're still puzzled about.
2
#2
2 Frags +

This would be brilliant

This would be brilliant
3
#3
43 Frags +

mike + 5

mike + 5
4
#4
10 Frags +

Sounds like a great idea!

Just something to consider is allowing people to choose 7 players instead of limiting it at 6, so that one player is a sub/merc (6 points scoring players and 1 sub).
The sub player only scores points when someone in the starting 6 doesn't play one week.
Therefore, people wont miss out on points just because a player they chose can't play one week due to IRL commitments.

Sounds like a great idea!

Just something to consider is allowing people to choose 7 players instead of limiting it at 6, so that one player is a sub/merc (6 points scoring players and 1 sub).
The sub player only scores points when someone in the starting 6 doesn't play one week.
Therefore, people wont miss out on points just because a player they chose can't play one week due to IRL commitments.
5
#5
10 Frags +

Also how are you going to calculate players values if there is no class specification? Like zebbosai on scout would have a completley different value in terms of $/points earned than he would on medic.

Also how are you going to calculate players values if there is no class specification? Like zebbosai on scout would have a completley different value in terms of $/points earned than he would on medic.
6
#6
6 Frags +

This would be very interesting, I can imagine it being a bit broken for the beginning but it'd be cool to see it working and would be fun.

This would be very interesting, I can imagine it being a bit broken for the beginning but it'd be cool to see it working and would be fun.
7
#7
-1 Frags +

why not play it. After there isn't a whole lot of players would be worth picking since there's so few players (if i compare this to sc2 proleague for example)

why not play it. After there isn't a whole lot of players would be worth picking since there's so few players (if i compare this to sc2 proleague for example)
8
#8
2 Frags +

How comes everything that implies GentlemanJon is amazing

are you a prophet

How comes everything that implies GentlemanJon is amazing

are you a prophet
9
#9
6 Frags +
kaidusAlso how are you going to calculate players values if there is no class specification? Like zebbosai on scout would have a completley different value in terms of $/points earned than he would on medic.

My thinking is that it essentially becomes the fantasy player's risk to take. You can buy Zebbosai, but have no way of knowing whether he's going to play scout or medic or fold the team. I'd try to price him at whatever it seems like he's playing at the start of the season and he may either becomes a fantastic bargain or a dead weight depending on what changes.

One idea I was toying with was a market mechanic so player values go up and down depending on their popularity but that makes it hard for latecomers to pick a top player, it could be something I only activate after a signup period, but it's complex and that would push back release a lot. Be an interesting element if I could get it right though.

[quote=kaidus]Also how are you going to calculate players values if there is no class specification? Like zebbosai on scout would have a completley different value in terms of $/points earned than he would on medic.[/quote]
My thinking is that it essentially becomes the fantasy player's risk to take. You can buy Zebbosai, but have no way of knowing whether he's going to play scout or medic or fold the team. I'd try to price him at whatever it seems like he's playing at the start of the season and he may either becomes a fantastic bargain or a dead weight depending on what changes.

One idea I was toying with was a market mechanic so player values go up and down depending on their popularity but that makes it hard for latecomers to pick a top player, it could be something I only activate after a signup period, but it's complex and that would push back release a lot. Be an interesting element if I could get it right though.
10
#10
1 Frags +

Bump. Anything to say from the (EU) day shift?

Bump. Anything to say from the (EU) day shift?
11
#11
7 Frags +

Another thing to do is to do it on a weekly basis, this does provide prem to actually schedule matches properly but you could give everyone a set budget for week 1, they pick players and the prices of said players either goes up or down based on their performance for that week. You can then base some extra bonus points based on their performance based on teams they are expected to win/lose against.

either way I like this idea.

Another thing to do is to do it on a weekly basis, this does provide prem to actually schedule matches properly but you could give everyone a set budget for week 1, they pick players and the prices of said players either goes up or down based on their performance for that week. You can then base some extra bonus points based on their performance based on teams they are expected to win/lose against.

either way I like this idea.
12
#12
0 Frags +

Is it going to be like team making in FIFA games? Or just a bit similar

Is it going to be like team making in FIFA games? Or just a bit similar
13
#13
8 Frags +
HerganIs it going to be like team making in FIFA games? Or just a bit similar

That would be amazing, I want my Mike Team of the Week Foil card

[quote=Hergan]Is it going to be like team making in FIFA games? Or just a bit similar[/quote]
That would be amazing, I want my Mike Team of the Week Foil card
14
#14
3 Frags +
HerganIs it going to be like team making in FIFA games? Or just a bit similar

My budget won't be quite as high as EAs, but as close as I can manage

[quote=Hergan]Is it going to be like team making in FIFA games? Or just a bit similar[/quote]
My budget won't be quite as high as EAs, but as close as I can manage
15
#15
2 Frags +

I always thought this would be fun, but I worried that the homogeneity of stats would make different classes meaningless to compare

therefore, even though high level 6s is more competitive, I once made a fantasy ruleset for highlander

http://pastebin.com/vE8Uqj3b

the system is far from perfect (pyros kept getting too many points), but here it is if it helps

I always thought this would be fun, but I worried that the homogeneity of stats would make different classes meaningless to compare

therefore, even though high level 6s is more competitive, I once made a fantasy ruleset for highlander

http://pastebin.com/vE8Uqj3b

the system is far from perfect (pyros kept getting too many points), but here it is if it helps
16
#16
18 Frags +

This is such a sick idea. I hope this takes off. I've wanted to do something like this for years but never had time to try and figure out a proper method for the scoring/drafting/etc.

IMO It seems like using a system like Fanduel's would be pretty perfect. Depending on which method you use for roster building I think you could do it one of two ways:

1. Essentially just as you said you get a budget that you must use to purchase your roster. The set up they use for fantasy basketball is 2 PG, 2SG, 2SF, 2PF, 1C, making 9 slots.

http://i.imgur.com/gRgpRms.jpg

or

2. If you use fantasy football. For football the lineup is: 1 QB, 2 WR, 2 RB, 1 TE, 1 W/R/T, 1 K, AND 1 DEF, again 9 slots.

http://i.imgur.com/R7YvCZ9.jpg

My thought process for the utility class is reserved for any plays made while NOT on any of the standard 6's classes. So basically its an extra slot for scout and roamers. And for utility you would take whatever cost for that player is and slash it in half, since they will hardly be using the utility in comparison to their normal class. For team, you make it similar to defense where winning/losing rounds affects the score, plus things like total domination or total frags, damage, this adds slight depth to the team slot as you might not want a team that will put up a 5-0 in 5 minutes as the frags/damage would be relatively low and therefore might not net as many points as a team that wins a long 5-4 or something along those lines.

You could probably change a lot of things with what I put here but these were just some crude ideas of possible roster methods.

This is such a sick idea. I hope this takes off. I've wanted to do something like this for years but never had time to try and figure out a proper method for the scoring/drafting/etc.

IMO It seems like using a system like Fanduel's would be pretty perfect. Depending on which method you use for roster building I think you could do it one of two ways:

1. Essentially just as you said you get a budget that you must use to purchase your roster. The set up they use for fantasy basketball is 2 PG, 2SG, 2SF, 2PF, 1C, making 9 slots.
[img]http://i.imgur.com/gRgpRms.jpg[/img]

or

2. If you use fantasy football. For football the lineup is: 1 QB, 2 WR, 2 RB, 1 TE, 1 W/R/T, 1 K, AND 1 DEF, again 9 slots.
[img]http://i.imgur.com/R7YvCZ9.jpg[/img]


My thought process for the utility class is reserved for any plays made while NOT on any of the standard 6's classes. So basically its an extra slot for scout and roamers. And for utility you would take whatever cost for that player is and slash it in half, since they will hardly be using the utility in comparison to their normal class. For team, you make it similar to defense where winning/losing rounds affects the score, plus things like total domination or total frags, damage, this adds slight depth to the team slot as you might not want a team that will put up a 5-0 in 5 minutes as the frags/damage would be relatively low and therefore might not net as many points as a team that wins a long 5-4 or something along those lines.

You could probably change a lot of things with what I put here but these were just some crude ideas of possible roster methods.
17
#17
0 Frags +

I really love the idea, i've been into fantasy football (NFL) for a while and also wanted to enter the new LoL fantasy stuff. I dont have much time to think about criticism or special ideas but i would advise you to have a look at vulcun(.com) they are the best website for fantasy LoL and it may give you an idea or 2 for the class issues concidering that league has 5 different roles to play. Also the system with the paid league and free leagues seems interessting although we obviously lack the manpower for a lot of paid leagues with high prices.
Anyhow, such an awesome idea !

I really love the idea, i've been into fantasy football (NFL) for a while and also wanted to enter the new LoL fantasy stuff. I dont have much time to think about criticism or special ideas but i would advise you to have a look at vulcun(.com) they are the best website for fantasy LoL and it may give you an idea or 2 for the class issues concidering that league has 5 different roles to play. Also the system with the paid league and free leagues seems interessting although we obviously lack the manpower for a lot of paid leagues with high prices.
Anyhow, such an awesome idea !
18
#18
8 Frags +

Let's talk about scoring

This is still in development, if I make a good run at it I might have something ready for the upcoming ESEA/ETF2L seasons.

At the moment I'm looking at something like this for each player:
+3 points for being on the winning team
+2 points per player for each round difference on the winning team
+1 for a kill
-1 for a death
+2 for popping uber
-3 for dropping uber
+1 for a medic kill
+2 kill by a medic
-1 for every other team member for every 5 times their medic dies
+2 for being top fragger
+2 for doing top damage
+2 for having the best KAD
+1 for longest killstreak
+1 for every 5 airshots
+1 for capturing a control point

I've had some suggestions for player of the week stuff but I think that would be a little complex to implement as a first step.

I'm not sure whether to put damage in there, something like a point for every 250 damage or so. The possibility of Conch spamming makes it a bit silly though. Is there anything else anyone would like to see? I'm trying to make sure it's workable from both ESEA stats which are pretty vanilla and a logs.tf page so anything too exotic probably won't be workable.

[b]Let's talk about scoring[/b]

This is still in development, if I make a good run at it I might have something ready for the upcoming ESEA/ETF2L seasons.

At the moment I'm looking at something like this for each player:
+3 points for being on the winning team
+2 points per player for each round difference on the winning team
+1 for a kill
-1 for a death
+2 for popping uber
-3 for dropping uber
+1 for a medic kill
+2 kill by a medic
-1 for every other team member for every 5 times their medic dies
+2 for being top fragger
+2 for doing top damage
+2 for having the best KAD
+1 for longest killstreak
+1 for every 5 airshots
+1 for capturing a control point

I've had some suggestions for player of the week stuff but I think that would be a little complex to implement as a first step.

I'm not sure whether to put damage in there, something like a point for every 250 damage or so. The possibility of Conch spamming makes it a bit silly though. Is there anything else anyone would like to see? I'm trying to make sure it's workable from both ESEA stats which are pretty vanilla and a logs.tf page so anything too exotic probably won't be workable.
19
#19
-23 Frags +

I would rather see you start with highlander, since it's harder to tell from basic logs or scoreboard who went big. Sentry repair or sandviches dispensed can make a huge diff in a hold

muh hilanderes BibleThump

I would rather see you start with highlander, since it's harder to tell from basic logs or scoreboard who went big. Sentry repair or sandviches dispensed can make a huge diff in a hold

muh hilanderes BibleThump
20
#20
-1 Frags +

there used to be one in NA but a lack of people using logs.tf I think killed it since the website was set up to parse logs.tf stuff. esp since it had stuff ss didnt like melee kills(citation needed)

there used to be one in NA but a lack of people using logs.tf I think killed it since the website was set up to parse logs.tf stuff. esp since it had stuff ss didnt like melee kills(citation needed)
21
#21
8 Frags +

I think an uber drop should take away at least as many points as popping an uber gives, if not more. The only issue I see with that is that kritz drops might fuck with scoring

I think an uber drop should take away at least as many points as popping an uber gives, if not more. The only issue I see with that is that kritz drops might fuck with scoring
22
#22
-2 Frags +

I would totally pay up to 10 or so USD to participate in this. Maybe a perk system in which your "budget" is determined by the amount you pay in? I don't know if that would be quite acceptable though.

I would totally pay up to 10 or so USD to participate in this. Maybe a perk system in which your "budget" is determined by the amount you pay in? I don't know if that would be quite acceptable though.
23
#23
3 Frags +
deetrI think an uber drop should take away at least as many points as popping an uber gives, if not more. The only issue I see with that is that kritz drops might fuck with scoring

OK, dropping updated to brutally punish medics for their failure

[quote=deetr]I think an uber drop should take away at least as many points as popping an uber gives, if not more. The only issue I see with that is that kritz drops might fuck with scoring[/quote]
OK, dropping updated to brutally punish medics for their failure
24
#24
0 Frags +
DontI would totally pay up to 10 or so USD to participate in this. Maybe a perk system in which your "budget" is determined by the amount you pay in? I don't know if that would be quite acceptable though.

I wouldn't want to open up pay-to-win options, but it would be nice if there were ways to support the site/generate some prizes. I'll have to give it some thought and try out some concrete ideas.

[quote=Dont]I would totally pay up to 10 or so USD to participate in this. Maybe a perk system in which your "budget" is determined by the amount you pay in? I don't know if that would be quite acceptable though.[/quote]
I wouldn't want to open up pay-to-win options, but it would be nice if there were ways to support the site/generate some prizes. I'll have to give it some thought and try out some concrete ideas.
25
#25
0 Frags +

Pay a key/arbitrary amount to enter, win and get them all/most of them and then use the leftovers to support the site

Pay a key/arbitrary amount to enter, win and get them all/most of them and then use the leftovers to support the site
26
#26
3 Frags +

I'm curious to hear what your thoughts are on the whole "people who play longer get higher stats" thing. Because I know it can become an issue based on how the scoring ends up being done. A brief experiment with fantasy stats was tried in UGC and this is what was found.. whoever played the longest games got the best stats.

And to use a real world example.. because ESEA is fresh in my head. Looking, for example, at something like most frags. Members of Mad Men topped this list - it goes YouMustMike, seymour, and then some others. You don't even see a Froyotech member until 12th. You see a similar thing with Damage, with three Mad Men members in the top 5. This is nothing against Mad Men - they were one of the top 4 NA teams this season. But should they be the Fantasy All-Star team just because they played at least an hour more of game time than everyone else? Shade only popped 112 ubers to indust's 233 - does that mean indust is twice as good as Shade? (All jokes aside?)

I am not criticizing, because I know you're a smart dude. I'd just like to hear you talk about how this can be tackled a little more, preferably without it just turning into a big game of comparisons to real sports. In my mind this is the toughest obstacle.

I'm curious to hear what your thoughts are on the whole "people who play longer get higher stats" thing. Because I know it can become an issue based on how the scoring ends up being done. A brief experiment with fantasy stats was tried in UGC and this is what was found.. whoever played the longest games got the best stats.

And to use a real world example.. because ESEA is fresh in my head. Looking, for example, at something like most frags. Members of Mad Men topped this list - it goes YouMustMike, seymour, and then some others. You don't even see a Froyotech member until 12th. You see a similar thing with Damage, with three Mad Men members in the top 5. This is nothing against Mad Men - they were one of the top 4 NA teams this season. But should they be the Fantasy All-Star team just because they played at least an hour more of game time than everyone else? Shade only popped 112 ubers to indust's 233 - does that mean indust is twice as good as Shade? (All jokes aside?)

I am not criticizing, because I know you're a smart dude. I'd just like to hear you talk about how this can be tackled a little more, preferably without it just turning into a big game of comparisons to real sports. In my mind this is the toughest obstacle.
27
#27
0 Frags +

Very nice idea, i've never did anything even close to this so i am really looking forward to it. For the sign up part i agree with brooky12

Very nice idea, i've never did anything even close to this so i am really looking forward to it. For the sign up part i agree with brooky12
28
#28
1 Frags +
GetawhaleI'm curious to hear what your thoughts are on the whole "people who play longer get higher stats" thing.

This reminds me of something I wanted to include, bonus points for crushing the opposition by a huge margin (although this creates a pain with payload when it comes to HL). It is a problem, and even including negative stats like deaths cancelling frag points doesn't completely mitigate it.

I have come up with some alternatives but they all have their flaws. The biggest problem with them is that they aren't intuitive, you can't just look at the log or match page or whatever and work it out for yourself, you need to get a spreadsheet out or something which divorces people from the proceedings.

I think looking at things that blanket reward teams for a fast and easy win is probably the approach that makes the most sense to players. I'm not sure where that point is, maybe 2 or 3 points per player for each round difference to the winning team.

It does bring up some other problems, teams dropping, logs failing, etc, but in those cases I think you just have to accept shit happens.

[quote=Getawhale]I'm curious to hear what your thoughts are on the whole "people who play longer get higher stats" thing. [/quote]
This reminds me of something I wanted to include, bonus points for crushing the opposition by a huge margin (although this creates a pain with payload when it comes to HL). It is a problem, and even including negative stats like deaths cancelling frag points doesn't completely mitigate it.

I have come up with some alternatives but they all have their flaws. The biggest problem with them is that they aren't intuitive, you can't just look at the log or match page or whatever and work it out for yourself, you need to get a spreadsheet out or something which divorces people from the proceedings.

I think looking at things that blanket reward teams for a fast and easy win is probably the approach that makes the most sense to players. I'm not sure where that point is, maybe 2 or 3 points per player for each round difference to the winning team.

It does bring up some other problems, teams dropping, logs failing, etc, but in those cases I think you just have to accept shit happens.
29
#29
0 Frags +
GetawhaleI'm curious to hear what your thoughts are on the whole "people who play longer get higher stats" thing. Because I know it can become an issue based on how the scoring ends up being done. A brief experiment with fantasy stats was tried in UGC and this is what was found.. whoever played the longest games got the best stats.

And to use a real world example.. because ESEA is fresh in my head. Looking, for example, at something like most frags. Members of Mad Men topped this list - it goes YouMustMike, seymour, and then some others. You don't even see a Froyotech member until 12th. You see a similar thing with Damage, with three Mad Men members in the top 5. This is nothing against Mad Men - they were one of the top 4 NA teams this season. But should they be the Fantasy All-Star team just because they played at least an hour more of game time than everyone else? Shade only popped 112 ubers to indust's 233 - does that mean indust is twice as good as Shade? (All jokes aside?)

I am not criticizing, because I know you're a smart dude. I'd just like to hear you talk about how this can be tackled a little more, preferably without it just turning into a big game of comparisons to real sports. In my mind this is the toughest obstacle.

frags per minute is an important stat when looking at an overall comparison

[quote=Getawhale]I'm curious to hear what your thoughts are on the whole "people who play longer get higher stats" thing. Because I know it can become an issue based on how the scoring ends up being done. A brief experiment with fantasy stats was tried in UGC and this is what was found.. whoever played the longest games got the best stats.

And to use a real world example.. because ESEA is fresh in my head. Looking, for example, at something like most frags. Members of Mad Men topped this list - it goes YouMustMike, seymour, and then some others. You don't even see a Froyotech member until 12th. You see a similar thing with Damage, with three Mad Men members in the top 5. This is nothing against Mad Men - they were one of the top 4 NA teams this season. But should they be the Fantasy All-Star team just because they played at least an hour more of game time than everyone else? Shade only popped 112 ubers to indust's 233 - does that mean indust is twice as good as Shade? (All jokes aside?)

I am not criticizing, because I know you're a smart dude. I'd just like to hear you talk about how this can be tackled a little more, preferably without it just turning into a big game of comparisons to real sports. In my mind this is the toughest obstacle.[/quote]
frags per minute is an important stat when looking at an overall comparison
30
#30
-1 Frags +

the biggest trick is balancing the scoring system so that every class is as useful as every other class. as i see it right now the ideal team would be 6 pockets, since they are likely to topfrag and damage every game, as well as have the highest FA/D. Some things to bring into account might be frags/damage, heals taken, and medic assists.

GetawhaleI'm curious to hear what your thoughts are on the whole "people who play longer get higher stats" thing. Because I know it can become an issue based on how the scoring ends up being done. A brief experiment with fantasy stats was tried in UGC and this is what was found.. whoever played the longest games got the best stats.

obviously everything will have to be done as a ratio (frags per death, damage per minute)

the biggest trick is balancing the scoring system so that every class is as useful as every other class. as i see it right now the ideal team would be 6 pockets, since they are likely to topfrag and damage every game, as well as have the highest FA/D. Some things to bring into account might be frags/damage, heals taken, and medic assists.

[quote=Getawhale]I'm curious to hear what your thoughts are on the whole "people who play longer get higher stats" thing. Because I know it can become an issue based on how the scoring ends up being done. A brief experiment with fantasy stats was tried in UGC and this is what was found.. whoever played the longest games got the best stats.[/quote]

obviously everything will have to be done as a ratio (frags per death, damage per minute)
1 2 3 4 ⋅⋅ 22
Please sign in through STEAM to post a comment.