The rules are a bit complicated (at least in my opinion):

https://www.whatifsports.com/knowledgebase/KB_Article_Details.aspx?kbid=674

Starters will be determined using the following rules:

• The formation that is most used in the game will determine the starting 11 positions
• The players at each of those 11 positions with the most number of plays will be picked as the starter at the end of each game
  • In the case where they have the same number of plays, the player on the default depth chart will get the nod as the starter

Additionally, any player that has 60% or greater playing time will also get credit for a start. This may cause an additional starter or two.

From Dev Chat:

The only real value that I see using a "starter" tag is its value in recruiting and making that promise. How will your next update affect that "promise"? (harriswb3 - Hall of Famer - 7:09 PM)

Currently the starter tag is still used for recruiting, the starters are calculated from the plays played and the depth charts. In the 3.1 release we are investigating ways to better improve this. The most probable way to go about this is to get rid of that promise in recruiting. It's value is similar to playing time for the recruit so eliminating it would not impact any decisions of a recruit.

Talking about Paul Rogers? Yeah, that is a weird situation. No clue what is going on.

Could be a deal where he got 50% one game then 20 and 20. Averages out to 30 but he only got enough snaps one game.

Hi ebel331,

There is an explanation as to how Paul Rogers is earning starts with limited playing time. It is very complex and difficult to explain, so if you want to send in a customer support ticket, please feel free. It is based on a combination of your playbook settings, formation settings and depth chart settings.

As stated above by nitros, you have a defensive formation that is used the most determined as your starting formation. The players who earn starts will then be based on the highest playing percentage at each position. If there are multiple players with the same amount playing time, they could each earn starts, meaning that you could have more than 11 players earn a start for a game.

Thank you!

Who is this WIS_Support guy? I see no NC's on his resume, so obviously a hack.

nitros

P.S. Just kidding. Thanks for jumping in on the thread!

Yes, thank you Support. I've always wondered about this too.

Hey WIS_Support. Since you're suggesting that ebel331 open a support ticket to get an explanation, can you provide an ETA on when the support ticket I opened on November 22nd will be answered?

I know this is a WIS support explanation. But the evidence gained from inspection of any experienced coaches game stats shows that this is not accurate.

Exhibit 1: Houston's QB

James Griffith

Chris Diaz

I run 3-4 defense every play so formation calculations are out.

I have a DL that got credit for a start with 32% played, but did not get credit with 36% played.

So that range appears to be the bubble where the 'complex and difficult to explain' comes in.

Posted by mraston on 1/19/2017 3:58:00 PM (view original):
I know this is a WIS support explanation. But the evidence gained from inspection of any experienced coaches game stats shows that this is not accurate.

Exhibit 1: Houston's QB

James Griffith

Chris Diaz

This is par for the course. The people supporting and, in fact, creating this game don't know how it actually works. Been this way since I can remember.

With respect to how Gridiron Dynasty actually works (I have no clue about the other games), a coach would really be better served shotgunning any questions out to 5-10 of the best/most-experienced coaches and seeing if anyone knows the answer. Or at least if a consensus could be developed. You'd get a better answer than asking Support.

In this case, mraston is right - the actually assignment of starts does not adhere to anything like what Support just told us. So either:

1. Support is mistaken in its impression of how Starts are assigned, OR
2. The way it's supposed to work is how it was explained, but the logic was coded so horribly that it doesn't actually work that way

Either way, I'm not sure it really matters WHY it doesn't work that way, so much as it matters HOW it actually works.