As the state-run Metropolitan Transportation Authority launched negotiations with the Transport Workers Union last year for a new three-year labor agreement, one of the authority’s biggest goals was to win the right to run “one-person trains” — that is, trains with an engineer rather than an engineer and a conductor — on two lines, the L and the #7 trains.

One-person train operation — Opto — has been a goal for years.

By late last year, the TWU had tentatively agreed to this deal, for a few things in return — like a $2 premium for engineers on the trains — that the MTA considered reasonable.

Yet as negotiations turned to arbitration and as arbitration neared its end, the MTA unilaterally yanked its request for Opto, seemingly snatching defeat from the jaws of victory.

Why?

The MTA’s argument, in a 6/19 memo from its lawyers to arbitrators and then later in the press, is two-fold.

First, months before sending its contract to state arbitration, the authority had agreed to cap workers’ healthcare contributions to 1.5 percent of straight 40-hour wages, rather than continue to apply the levy to 40 hours plus overtime.

But earlier this year, the MTA seemed to have had a change of heart, deciding it wanted to keep the higher healthcare charge. It clumsily tried to give up Opto in return (and ended up losing on both counts).

Second, the MTA’s lawyers said in the memo that with respect to Opto, that there was no understanding between the parties as to the details of the deal, with regard, in particular, to the union possibly objecting to safety conditions after signing the deal, and throwing it into limbo.

Later, the MTA argued that Opto had become a cost rather than a savings, because the union had upped its premium wage required for engineers riding solo to $4 rather than $2, and because Opto required the MTA to hire 40 new supervisors (supervisors that the arbitrators, strangely, awarded even without Opto).

Do these arguments add up?

The safety argument seems tenuous. For decades, the TWU has had the right to throw any matter into emergency arbitration if it sees a risk to health or safety, and such arbitration is supposed to result in a quick decision.

So, sure, it’s possible that the TWU could have thrown Opto into safety arbitration after agreeing to it — and then the union would have either won, or lost.

If the union lost, the MTA would have had Opto. If the union won, the MTA would have lost Opto … so instead, to prevent this outcome, the MTA unilaterally gave Opto up … ???

As for the cost: purely in terms of Opto being a short-term cost rather than a savings, the MTA may be correct.

But this shows not good management but shortsightedness, an MTA specialty.

Consider: during phase-in, the MTA could have been in a situation where it was paying engineers $4 more while still having a surplus of conductors.

But over time, paying engineers $4 more outweighs paying conductors $21 an hour, and having to fund 40 supervisors vs. employ potentially hundreds of conductors should do so, too.

There’s a reason why right now was the best time for the MTA to incur this short-term cost.

For various strange political reasons, the TWU is likely more willing than it ever will be to do something that hurts conductors.

In a few years, that may not be the case.

The MTA could have used this window of opportunity to show engineers — and conductors who want to become engineers — that this change was a good thing for them, because it would have created desired jobs that paid a little bit more — thus encouraging the TWU to want to expand the deal a few years hence.

It could have used this time to see how Opto actually works in practice when deployed beyond its previous experiment on the L line — working out what kind of safety provisions it needs to make, for example, in whether it can use Opto at rush hours without sacrificing the ability to evacuate trains quickly, and using some resources for new police patrols to replace the conductor’s passive safety presence with an active one.

Eventually, if Opto worked well and with more investment in technology, the MTA could have won permission to expand Opto to all lines, at least during some hours of the day, potentially, gradually cutting train-operation costs nearly in half.

Now, the opportunity to put a foot in the door here is gone.

At best, killing Opto may show that the MTA head office, especially without clear leadership, is suffering under a conventional-wisdom tyranny — fearful of change unless the results are guaranteed.

This hurts, when the authority’s biggest challenge is labor flexibility.

The State Senate will soon hold hearings on Paterson’s pick to be the next MTA chairman and chief executive, Jay Walder.

If it emerges from its corrupt torpor, the Senate should extensively question Walder on his views on labor flexibility — and ask him to bone up on the details of this episode in advance, so that he can offer an opinion here, too.