The operative words here are "My best guess"
Nearly all equity splits are guesses. The CEO and CTO guessed once (probably 50/50) and now something has changed (a new person) and they realize their initial guess was wrong and have to guess again. This pattern means every time something changes the equity split will have to be renegotiated. This gets harder and harder.
You don't need a number, you need a formula. Slicing Pie is the only formula on the planet that will allow you to determine exactly how much equity each person deserves no matter what changes or who joins the team or who leaves.
Let's pretend your company was a hand of Blackjack. The CTO and the CEO agree to go in 50/50 and each put $1 on the same hand. If they win, they would split the winnings. But not everything goes as planned. The dealer deals two aces and the game gets more interesting. They decide to split the aces and double-down. The CEO puts down $1, but the CTO is broke. So, they ask this new guy to play. He comes on board and bets $1. So, now the CEO has bet $2, the CTO and #3 have each bet $1. The 50/50 split no longer works. It should be obvious that the split should be CEO: 50% (he bet $2), CTO 25% (he bet $1) and #3: 25% (he also bet $1).
What's going on here is that over time people are making new bets. Each time they bet the old split makes no sense. You need an equity model that can self-adjust to changes.
When people contribute time, money, ideas, relationships, supplies, equipment or anything else to a startup they are essentially betting on the future outcome of the startup. The amount of their bet is equal to the fair market value of their contribution.
A person's % share of the equity should always equal that person's % share of the risk.
Slicing Pie is a formula that accounts for each person's risk and allocates equity accordingly. It also has a formula for recovering equity in the event that someone leaves.
I've written a book about how this all works and you may have a copy if you contact me through SlicingPie.com