React Strictmode preventing a custom made useState hook to change the state at the first call

Question

I was following along a presentation of Ryan florence (creator of remix and ReactTraining.com). There he was demistifying useState hook, by making one of his own.
I followed along the procedure, and implemented this in a component.

import React from 'react'
import ReactDOM from 'react-dom';



function MadeUseStateMyself() {
  const [minutes,setMinutes] = useState(5)
  const [error,setError] = useState(null)
  
  const handleAdd = () => {
      if(minutes<9){
          setMinutes(minutes+1)
          setError(null)
      }
      else{
          setError('Less than 10 please')
      }
  }
  const handleSubStract = () => {
    if(minutes>0){
        setMinutes(minutes-1)
        setError(null)
    }
    else{
        setError('More than 0 please')
    }
}

  return (
    <React.Fragment>
        <button onClick={handleSubStract} > - </button>
        <p>{minutes}</p>
        <button onClick={handleAdd} > + </button>
        {error && <div>{error}</div>}
    </React.Fragment>
  )
}
const states = []
let calls = -1
const useState = def => {
    const callId = ++calls
    if(states[callId]) {
        return states[callId]
    }
    const setDef = newV => {
       states[callId][0] = newV
       reRender()
    }
    const state = [def,setDef]
    states.push(state)
    return state
}
function reRender(){
    calls = -1
    ReactDOM.render(<MadeUseStateMyself/>,document.getElementById('root'))
}
export default MadeUseStateMyself

Here I noticed a really peculiar behavior, that is, whenever I use React.StrictMode wrapping my entire component, the state doesnt change after the first function call(In this case, I implemented an incremental and a decremental button which changes the state of one singular integer state.)
This problem doesnt rise if I get rid of the React.StrictMode wrapper. Here is a CodeSandbox

Reading the StrictMode docs , one of my assumptions is that as in strictmode, function component bodies are invoked twice, it loses its first state during the second function call, thus losing the first change of state.
Am I even close?