What exactly does he 'shut down'?
He performs quite poorly against the other teams 2nd lines. When he is on the ice the other team scores more goals than we do by a significant margin (and he mostly plays against 2nd comp).
When Nurse-Bear are on the ice, they typically outscore the opposition (who are top comp most of the time, mind you).
Ethan Bear is the best 'shutdown' defenceman we have, because he limits the amount of chances the opposition has on net more than anyone else on our team. Larsson wishes he was half the shutdown defenceman Bear is. I know Lars fits the stereotypical mold of 'shutdown' defender (big, slow, bad with the puck) but this doesn't mean he is conducive to winning hockey games. In fact, stats show he's the opposite.
Larsson has been very good for well over a month now. To me he looks a lot more like his old self. And statistically this shows as well. Since Jan 1 here are some of the numbers you seem to be referencing. I choose Jan 1 since that is when the team came out of its disastrous run. If you go by the whole season Larsson's numbers are skewed by a higher portion of his minutes being played during the period where the whole team was getting destroyed.
These are 5 vs 5 numbers from Natural Stats trick.
Larsson 299:28 TOI CA/60 53.5 CF% 50.00 GA/60 2.4 GF% 50% SCA/60 23.84 SCF% 49.79 HDCA/60 9.43 HDCF% 53.00
Nurse 343:40 TOI CA/60 57.79 CF% 50.67 GA/60 2.27 GF% 56.67% SCA/60 26.36 SCF% 48.87 HDCA/60 10.82 HDCF% 47.01
Nurse has the better GF% but the others stats are actually fairly comparable.
Of course these numbers are not really individual measures of a player's ability but they do show that in general when Larsson has been on the ice the team has fared well since the New Year.
I would not call Larsson untouchable. But I don't think I'd trade him unless the deal was a very good one. His ability to break up the cycle is something that I think is more valuable than it seems you do. I also think he has more offensive ability in the oppositions end than you give him credit for.