First Player Advantage: Possible Fix

RCubed1967

I have been collecting data for the past 3 months for every game I play in the Rubedite level. I have tried multiple strategies, yet the first pick position always has a statistically better chance of winning. The table below shows the results of my 638 ranked games (yes, my win rate sucked, but I have gotten better over time).



My data suggests that you have a 40% better chance of winning if you have first pick. By picking patrons that have fewer first pick advantages, a first mover advantage is reduced but still exists.

I think a simple fix is to give the second player a writ of coin in their hand on their first turn, replacing a coin rather than giving one extra gold. The writ of coin will provide a slight improvement over the entire game that should partially offset the huge first pick advantage.

I would love to see data that either agrees or disagrees with my analysis on first move advantage (data...not gut instinct).

AnduinTryggva

I also made a similar statistics a while back and I came to a similar result.

But I have some difficulty to understand the difference between getting an extra gold during the first turn from your solution which is, as far as I understand, giving the second player a card during the first turn that gives him/her also gold (presumably 1). Can you elaborate a bit more your idea?

FrancisCrawford

AnduinTryggva wrote: »

I also made a similar statistics a while back and I came to a similar result.

But I have some difficulty to understand the difference between getting an extra gold during the first turn from your solution which is, as far as I understand, giving the second player a card during the first turn that gives him/her also gold (presumably 1). Can you elaborate a bit more your idea?

If I understand correctly, the second player could exit the first turn with 2 Writs of Coin rather than 1.

AnduinTryggva

So in the next turn the second player has 2 additional gold? Or how is this to be understood? I am still lost.

BTW here's my data. Not as large a sample as that of OP but still similar tendencies.

https://forums.elderscrollsonline.com/en/discussion/612592/starting-drawing-clear-advantage-as-shown-by-my-data/p1

RCubed1967

Awesome.

To clarify, we start with 6 coins and 4 Patron cards. If we chose to upgrade a coin, it becomes a writ of coin. The writ of coin is worth 2 gold.

The second pick player has one their six coins already upgraded to a writ of coin. That writ of coin is automatically in the second players initial draw of 5 cards and replaces the extra gold currently in the game. In effect, player 2 stands exactly where they stood on turn 1 as before. The writ of gold only comes into play as the game goes on.

The writ of coin is shuffled into the discard pile with the other cards at the end of the turn. When they see the writ of coin again is up to the god of chance. The only way a player could get the writ of coin again on turn 2 is if they drew an extra card and the god of chance liked them. Most likely, the player would not see that writ of gold again until turn 3, 4 or 5.

Your data was good. There was a comment made in your posts about choosing different patrons to counter first mover advantage. I did this and there is an improvement, but it does not eliminate the first turn advantage.

I think the writ of coin idea is enough to tweak the game chance to even the odds (with smart play of course).

NoSoup

One thing the statisitcs can't tell us is where the advantage is coming from. Is it from having the first pick from the tavern or does the advantage come from something earlier.

The person with the first tavern pick also gets the first Patron selection. This is a huge advantage because as I might start the game with an idea of which two patrons I pick, I quite often change it to counter/better suit the 2 patrons the other players pick. Personnally, I think the only thing that needs to change is flipping who goes first. The person who has to pick the patrons second gets the first go at the tavern. Having a gold coin early is already a pretty good advantage for going second.

spartaxoxo

I do think being able to counter pick a deck based off player two's choices probably does have an impact on winning. The other day I picked Orgnum after they picked Rahjiin purely because I was hoping it would make them hesitate to give me Bewilderment cards because they'd try to keep up with Orgnum spam. It worked and I won.

RCubed1967

NoSoup - a few points on this.

This is exactly where statistics tells us the issue is. I had a large sample size. Another person's independent data agreed with the results. I varied my strategy in the second pick position to try to improve it (again with large sample sizes). This is the realm of statistics and is therefore valid.

If I had not varied my second start position, then we have to consider alternative possibilities. Further, if I did not try to improve my second pick results, I would be following Einstein's definition of insanity: doing the same thing over and over and expecting different results.

Flipping might be an issue, but does not remove the first turn advantage. In the 638 game sample I provided, I got first pick 52.2% of the time (probability of occurring is 1.71%, therefore significant because is it less than 5%). The game made up for itself at the start of this month though...my first 10 games in Rubedite were second pick. 0/10 = 0.09766% or 1 in 1024. I should never see that occur again unless I play >10,000 ranked games. I will watch over time.

We could also address the first pick issue by working some card imbalances...but I think that is more controversial. For that, we would have to look at cards such as Armory.

Patron balancing is also possible, but I personally don't like the idea (EXCEPT THAT ALMALEXIA CONFINES NEED TAUNT). Picking patron imbalances is part of the strategy.

NoSoup

RCubed1967 wrote: »

NoSoup - a few points on this.

This is exactly where statistics tells us the issue is. I had a large sample size. Another person's independent data agreed with the results. I varied my strategy in the second pick position to try to improve it (again with large sample sizes). This is the realm of statistics and is therefore valid.

If I had not varied my second start position, then we have to consider alternative possibilities. Further, if I did not try to improve my second pick results, I would be following Einstein's definition of insanity: doing the same thing over and over and expecting different results.

Flipping might be an issue, but does not remove the first turn advantage. In the 638 game sample I provided, I got first pick 52.2% of the time (probability of occurring is 1.71%, therefore significant because is it less than 5%). The game made up for itself at the start of this month though...my first 10 games in Rubedite were second pick. 0/10 = 0.09766% or 1 in 1024. I should never see that occur again unless I play >10,000 ranked games. I will watch over time.

We could also address the first pick issue by working some card imbalances...but I think that is more controversial. For that, we would have to look at cards such as Armory.

Patron balancing is also possible, but I personally don't like the idea (EXCEPT THAT ALMALEXIA CONFINES NEED TAUNT). Picking patron imbalances is part of the strategy.

Ummm. The problem is you can't verify whether the advantage comes from selecting the last patron or having the first turn as one player gets both these options. What you're saying about changing up your second picks only increases my agrument that a big part of the advantage comes from having the first and last pick. You changing up your second options does nothing because the player that goes first can still pick the patron that best suits their playstyle vs the 2 patrons you picked.

To rule out the effect of patron selection you would have to select four patrons and then only use the data from games played where those same four specific patrons are used.

RCubed1967

Rather than debating, additional data is the only thing which helps either argument. My July data so far indicates similar results.

I suggest that you grab a sample of games and confirm if you get the same results or not. Your sample needs to contain at least 60 games, with a minimum 30 first picks and a minimum 30 second picks. Do not filter any results for any reason (except make the games PvE in Rubedite).

If you want to make your sample bigger and more varied, please do so. Just make sure each condition you record has at least 30 second picks. 30 is important as it makes it statistically relevant. More data (games) is better!

Remember, we are looking to determine if your first pick has a higher win rate than your second pick.

I encourage others to do the same. The more data, the better.

Respectfully, I would like to see statistically sound data to support your side of the debate.

ZOS - if you are paying attention, you have all the data at your disposal. What is the win rate for first pick versus the win rate for second pick?

Necrotech_Master

the biggest advantage i see the first player having is first pick of the tavern

the first to play does not have the 1 extra coin, so only 5 coin to work with on the turn, so to get a writ of coin and buy a card in the same turn the card would have to cost 3 coin or less (and there are some good 3 coin cards that can make this work), or you could get 1 good 5 coin cost card, but your locked out of anything 6 coin or higher

the 2nd player to go gets the +1 bonus coin, with their initial hand is 6 coin to spend, so they can buy something of cost 4 and still make a writ of coin, or buy a single good card costing up to 6 (which really opens up as there are a lot of good 6 coin cost cards, such as armory if pelin is in play, the first player to move does not have access to buy that turn 1 due to costing 6 coin)

so, in a lot of cases, i actually almost prefer going 2nd and starting with the +1 coin because i feel i can actually have better options, but because of the RNG of the tavern, even 2 players making optimal moves, its really going to depend on what shows up in the tavern first turn more than who gets to go first

tsaescishoeshiner

I doubt there's any major way that first and last pick impact win ratio, because you're picking the decks that both players use. If my opponent picks Rajhin and I pick patrons that I believe counter it, I'm just as likely to pick up a Rajhin card as they are.

There are obviously ways that pick order could influence a game, but we won't get data to test it, and it doesn't strike me as the most likely area of imbalance. Picking the first card from the tavern, getting the first deck cycle, and having a headstart on income and power seem like more of an advantage.

Instead we have to theorize about solutions to give the player going second less of a disadvantage.

For example, what if the second player started with a contract card in their hand that said "Replace one card from the tavern." If that's too much, maybe there could be limitations or a downside added to it.

I wonder—what's the average amount of prestige that the first player wins by? Maybe the player who goes second could start with a few points of prestige.

AnduinTryggva

NoSoup wrote: »

One thing the statisitcs can't tell us is where the advantage is coming from. Is it from having the first pick from the tavern or does the advantage come from something earlier.

The person with the first tavern pick also gets the first Patron selection. This is a huge advantage because as I might start the game with an idea of which two patrons I pick, I quite often change it to counter/better suit the 2 patrons the other players pick. Personnally, I think the only thing that needs to change is flipping who goes first. The person who has to pick the patrons second gets the first go at the tavern. Having a gold coin early is already a pretty good advantage for going second.

Indeed, I also came to the conclusion that being the one to pick the last patron is a small advantage and it can be used in a tactically advantageous way.

It is ONE advantage. However, as both patrons are available to both players, the ability to pick a good card before the other is much more important, specifically during the first rounds.

Seraphayel

I really think giving the second player a Writ directly instead of an additional coin just for one round would solve a lot of issues in the long run. You basically spare yourself a turn to yourself exchange a coin for a Writ and therefore can make use of a patron in that round in the best case. That one coin can sometimes be nice when the tavern RNG allows it, but often it‘s a waste because you can’t make use of it. With a Writ this problem would be solved completely as you still benefit by it one way or another.