Let's renormalize the 1.4%

"1.4 per cent mortality for individuals subdued by police using a taser..." [LINK]

"An analysis of 183 Victoria police incident reports from 2005 to 2007 revealed the device was used in push-stun in 57 per cent of all cases where police activated their tasers." [LINK]

As Taser well knows, using the taser in push-stun (or 'Drive Mode', almost always not into the chest) is vastly safer (with respect to cardiac issues) than shooting the barbs and trailing wires towards the chest. That's why Taser loves to wash out the statistics (denominator washing) by including all deployments in any mode.

So right off the bat, we can make one correction to the overall mortality rate by adjusting out the Drive Mode deployments using the Victoria incident statistics given above.

1.4% times (inverse of (100%-57%)) = 3.3%

In other words, if the overall mortality rate for all individuals 'subdued by police using tasers' in any mode is 1.4%, and if 'Drive Mode' is generally much much safer, and if deploying the barbs and wires only occurs in about 43% of all usages, then the actual mortality rate when using the barbs and wires must be about 3.3%.

And this is being very generous due to lack of detailed data. For example, I suspect that there may be other categories of deployment such that my calculated 43% (from 100% - 57%) is too high. If so, then the calculated 3.3% mortality rate would be that much higher again.

And this is just the first correction.

Another correction might be required for the reported 20% failure rate [LINK] Or the reported 10% ineffective rate. [ibid]

3.3% becomes 4.1% becomes 4.6%.

Now we're getting within sight of the unwashed (no denominator washing) mortality rate for full-on X26 taser deployments where the barbs land on the chest.

And it all roughly matches: [LINK] [LINK] [LINK] [LINK]