Which Way Should I Go Now?

Part 2 of Evaluating Police Tactics: An Empirical Assessment of Room Entry Techniques

Here is the second part of the review.

The Publication

Evaluating Police Tactics: An Empirical Assessment of Room Entry Techniques by J. P. Blair and M. H. Martaindale. Published by Routledge in 2014.

What is the Issue?

There has been an argument in tactical policing for decades about what is the best way to enter a room that might have a hostile suspect in it. In this book, they look at two major issues regarding room entry with a focus on room entry during active attacks. This post reviews their research on the second issue.

The second issue addressed what direction the first officer who enters a room should go. For this report, they assume that a threshold evaluation has been done and that the door is in a corner of the room (referred to as a corner fed room). This means that if there is a hostile suspect in the room, he is likely in what is referred to as the hard corner of the room (the corner that is on the same wall as the door at the other side of the room – see illustration below).

There are three major schools of thought on which direction the officer should go. The first advocates that the first officer to enter should go on path A in the illustration. The second officer follows path B. This entry is referred to as Unknown because the first officer moves toward the corner that hasn’t been seen - so it is unknown. Because the first officer is moving directly toward the suspect, it is believed that this gives the first officer the fastest and most accurate shots on the suspect.

The second school argues that while moving toward the unknown corner might give the first officer the fastest and most accurate shots, it will also give the suspect more accurate shots on the officer. They also argue that because the suspect is already in the room and ready to shoot anyone, his shooting action will be faster than the first officer’s shooting reaction. This school also argues that because the second officer must move behind the first officer (line B in the illustration), the second officer will cross the suspect’s arc of fire for the first officer and that this might expose the second officer to gunfire intended for the first. To counter this, this school argues that the order of entry should be reversed. The first officer should move on line B, and the second on line A. Because the first officer goes toward a known (seen) corner in this entry, it is called Known.

Proponents of the Known entry argue that moving laterally should reduce the accuracy of the suspect shooting at the first officer. They also argue that as the suspect tracks the first officer, the suspect’s arc of fire shifts away from the door and this means that the second officer does not have to enter the arc of fire to move on path A. Proponents of the Unknown argue that the lateral movement will reduce the first officer’s accuracy. Proponents of the Known generally concede this point.    

The third school argues that they can get the benefits of both approaches by having the first officer enter on line C and the second on line A. Their argument is that the lateral movement on line C is enough to reduce suspect accuracy, but that it should not affect the accuracy or speed of fire of the first officer. Additionally, the lateral movement makes it so that the second officer can enter without crossing the suspect’s arc of fire at the first officer. This entry is called Hybrid because it splits the difference between the two to get the benefits of both.

How Did They Look at It?

The researchers conducted 4 studies to look at these issues.

Study 1

The first used 69 college students as the suspects. The students were placed in the hard corner of the room with a simunitions pistol loaded with 4 rounds, and told to shoot the officers when they entered. Then a team of two police officers was randomly assigned to conduct one of the three entries against the students. The researchers recorded reaction time and how many rounds hit each of the officers.

They found that the suspect’s reaction was generally faster than the officers who were helping with the study and the suspects were also faster than officers fired in other reaction time studies. The first officer to enter was hit more often during the Unknown entry (average of 1.94 hits) than in the Known (1.28 hits) or Hybrid (1.1 hits). This difference was statistically significant and the effect size was moderate. The differences in hits on the second officer by entry type were small and not statically significant. When the hits on both officers were combined, the difference in hits was statically significant with more hits on the officers in the Unknown condition (2.29 hits on average) than in the Known (1.61 hits) or Hybrid (1.57 hits). These differences were again moderate in size. The researchers also confirmed their suspicion that the second officer could be hit in the Unknown entries when passing behind the first officer. This occurred in about 16% of the Unknown runs.

Study 2

The second study was designed to look at first officer accuracy. Twenty-five licensed law enforcement officers and 3 military personnel who were attending an active shooter response conference participated. They were randomly paired with each other to create 14 teams. Each of these teams received training in how to perform each of the entries and then performed each of these entries during the experiment using a counter-balanced design to control for possible ordering effects. The suspect was a lab assistant placed in the hard corner of the room with instructions to start firing at the officers as soon as they entered. The officers each had 3 rounds to shoot at the suspect and these rounds were different colors so that they could tell which officer’s shots hit the suspect.

The primary outcome measure of this study was the accuracy of the first officer. In the Unknown condition, the average number of hits was 1.50. in the Known, it was 1.36, and in the Hybrid, it was 1.86. These differences were not statistically significant and suggested a small effect for entry type on officer accuracy.

Study 3

The third study was designed to look at first officer accuracy (again) and how long it took the second officer to fire. Ten members from a SWAT team were randomly paired with a partner creating 5 groups. Each of these groups then completed 6 room entries. Three entries were completed with one officer as the first officer and then the other officer completed three entries as the first officer. The first officer was given three simmunition rounds to shoot and the second officer was given a single round. The order of the entries was counterbalanced to control for ordering affects.

Again in this study, there was not a statistically significant difference for accuracy of the first officer for the different entries (Unknown = 1.8, Known = 2.2, Hybrid = 2.1). There was a statistically significant difference for how long it took the second officer to fire based on entry type. The second officer fired more quickly during the Known (1.29 seconds) and Hybrid (1.31 seconds) entries than in the unknown (1.61 seconds). This difference was only about 1/3 of a second, but in a real world gunfight (and statistics) that is a large difference!

The experimenters were surprised to find again that there was not a significant difference in accuracy by entry type. It is a common belief in firearms training that there is a speed/accuracy tradeoff in shooting. The faster someone shoots, the less accurate they are, and accuracy increases when shooters slow down. The experimenters posited that given that most of the participants in the studies had substantial firearms experience, they may have realized that the shots during the Known entry were more difficult and slowed their shots down. The experimenters went back and coded how long it took the first officer to fire during each of the entries in the third study. In the Unknown entry, the officer fired about .64 seconds after entry on average. In the Known entry, it took them about .77 seconds, and in the Hybrid, it took .70 seconds. These differences were moderate in statistical magnitude, but due to a violation of a statistical assumption, they were not statistically significant. Given that these differences were also not predicted before the study, the researchers decided to conduct another study to examine this issue.

Study 4

Fourteen officers from a SWAT team completed the fourth study. The procedure for this study was identical to the third study. This meant that 7 teams of officers completed 6 entries apiece with one officer being first for 3 entries and the other for the other 3. Entry order was again counter-balanced to control order effects.

Once again, the experimenters did not find a statistically significant difference in accuracy for the first officers by entry type (Unknown = 2.29. Known = 2.14, Hybrid = 2.50). When they examined how long it took officers to fire, they did find a statistically significant difference. It took officers in the Unknown entries an average of .89 seconds to fire. In the known group, it took 1.29 seconds, and in the Hybrid, it took .96 seconds. This suggested that the first officer fired faster during the unknown and hybrid entries than the Unknown. The magnitude of this difference was statistically moderate in size. Again, they found that the second officer was statistically significantly slower to fire in the Unknown entries (2.10 seconds) than in the Known (1.63 seconds) and Hybrid (1.70).

So What?

A summary of the statistically significant differences between entry type is presented below. Reading down the columns, a green box indicates that the entry was significantly better than the entry(ies) with red boxes. Reading across the chart, you can see that the Hybrid entry is the only one that has green boxes across all four categories. This suggests that overall, it is the better entry.

It is also worth noting, that shooting accuracy for the first officer was higher in the Hybrid condition than in the Unknown condition across all the studies. This difference was not statistically significant in any of the individual studies, but it was always in the same direction across studies. It might be that there is a small effect for accuracy between the two entries that these studies lacked the statistical power to detect (I’ll have to see if I can find the original data and see if there is a significant effect with the studies combined – If there is, I’ll post on it).

Most of differences in these studies are small in terms of time (often tenths of a second) and hits (often a third of a hit), but in terms of an actual gunfight, these differences can mean life or death. All other things equal, if I was performing an entry on a corner fed room that might have a hostile suspect, I would use the Hybrid. Of course, this is not possible with some room setups and there could be other reasons to do a different entry. I think that the core principle here is that adding some lateral movement relative to where the suspect is likely to be should reduce his accuracy in shooting at you, but you don’t want to do so much lateral movement that it slows your shots at him. It’s also a good thing to get your partner into the gunfight quickly and cleanly in case you are disabled when you enter.