Armor Metrics: Applying Lessons from the Statistical Revolution in Sports to Better Train Soldiers at the Company Level

Moneyball, a book by Michael Lewis about Billy Beane and the Oakland Athletics, hit the top of the New York Times best-seller list in 2003.1 Unable to compete financially with major market teams such as the New York Yankees and the Boston Red Sox, Beane and his staff used advanced statistics to increase efficiency in evaluating the effectiveness of players. Their evaluation tools produced results that questioned many of the long-held beliefs in professional baseball and sparked a statistical revolution that brought modernized and analytical performance measures from the periphery to the forefront of every managerial office. “Sabermetrics,” defined by founder Bill James as the “search for objective knowledge about baseball,” questioned the traditional measures of baseball skill and attempted to create new methods to better determine the value and efficiency of players.2 General managers no longer assessed players based on “baseball card” statistics such as batting average or runs batted in, but on statistics such as on-base plus slugging and runs scored, hence a much more efficient individual measure.3 Once the territory of “stat geeks” who resided in the periphery of professional sports, these statisticians have cemented their place in every major sport and contributed greatly to the success of professional teams. The task is clear: If statistical analysis is at the cutting edge of industry standards and proven to be a successful evaluation tool in corporate America, Wall Street and now professional sports, it is evident that Armor leaders must assess the lessons learned from statistical analysis and seek to apply methods to better train and evaluate their own units. This article serves to argue the use of statistical analysis in Armor companies. Using statistical analysis, Armor leaders can:

• Better assess the proficiency of Soldiers and tailor training according to their weaknesses;

• Provide leaders the tools necessary to best place and use Soldiers throughout their fighting force; and

• Create an environment of competition and esprit de corps that drives and motivates Soldiers to become the best at their given positions. Tragedy and opportunity As the U.S. Army’s only forward-deployed committed division, the 2nd Infantry Division’s posture to “fight tonight” is amplified and reinforced by recent events on the Korean Peninsula. A North Korean weapon system attacked and sank the Cheonan, a South Korean Navy ship carrying 104 personnel, off the country’s west coast March 26, 2010, killing 46 seamen.4 The bombardment of Yeonpyeong Nov. 23, 2010, put 2nd Infantry Division on its highest alert since the Korean War ended; the artillery engagement between the North Korean military and South Armor Metrics: Applying Lessons from the Statistical Revolution in Sports to Better Train Soldiers at the Company Level by CPT Michael B. Kim and SPC Mark S. Rothenmeyer

Korean forces resulted in two Republic of Korea Marines killed in action, two civilian deaths and 18 people wounded.5 The 2nd Infantry Division stands to deter North Korean aggression. Should deterrence fail, however, it trains to repel North Korean forces using conventional warfare. As such, this places greater focus on soldiers, noncommissioned officers and officers to become tactically and technically proficient on M1 Abrams tanks. The operating environment in the Middle East has shifted the training focus of Armor units to non-conventional and counterinsurgency core competencies (and rightly so). However, with the threat on the Korean Peninsula, 2nd Infantry Division Armor units have the opportunity to concentrate on the maneuvering and firing of M1 Abrams tanks. The challenge As company commander of Company C, 1st Battalion, 72nd Armor Regiment, the fielding of the M1A2 System-Enhancement Package tanks provided a unique opportunity to implement a gunnery training and evaluation system from the ground upward. I challenged my master gunners, SSG Zachary Siemers and SSG Donald Fermaint (who replaced SSG Siemers halfway through this trial), to help me use the lessons-learned from the statistical revolution in sports to develop new methods to augment the assessment tools provided in the 1st Heavy Brigade Combat Team gunnery manual (Field Manual 3-20.21). The operator new-equipment training process allowed us to create, implement and experiment with new assessment tools that would give us a better evaluation of each tank crewmember. Due to the complexity of the task and the limited time allotted during the OPNET process, assessing “gunners” became the priority of focus. Garnering lessons learned and trends from the statistical revolution in sports, we created “gunner statistics” to rate and evaluate individual gunners. The need As a company commander or company master gunner, many of our assessments of individual Soldiers are subjective. An evaluator is limited in the tools he can use to analyze and assess gunner performance. We simply assess a gunner’s value based on previous gunnery scores and subjective assessments. Ask a first sergeant or platoon sergeant who the top four gunners are in a company, and his answer is 80 percent subjective. It’s based on instinct or preference, not any analytic tools or evaluations. Perhaps he is accurate in his assessment, but that does not excuse the fact that there is not a systematic and objective approach in evaluating gunners. A gunner’s previous gunnery score has many additional variables that may not accurately reflect the gunner’s current capabilities. What about a company’s seventh or 10th best gunner? The company commander and master gunner do not currently possess the tools needed to evaluate each of their gunners. The method The primary challenge in this process is twofold:

• To determine evaluation criteria that best captures gunner proficiency; and

• To implement subjective assessments into an objective equation to account for crew chemistry and teamwork. Defining “gunner proficiency” turned out to be a learning process for the command team. It provided an opportunity to determine what an Armor company requires of its gunners. Wading through all the additional tasks and requirements asked of an M1A2 SEP gunner, we determined the following four variables were the most important:

• Speed in acquiring targets;

• Accuracy in hitting targets;

• Consistency in destroying targets; and

• Tactical/technical competency. The bottom line is that the gunner must obtain and destroy the enemy. The most challenging aspect in defining gunner proficiency is to quantify subjective variables such as team chemistry into the equation. In The Book of Basketball, Bill Simmons refers to chemistry in a team as “the secret.”6 Given to him by Hall of Famer Isaiah Thomas, the premise is that relationships and chemistry on a team contribute just as much, if not more, to an organization’s success than actual individual skill. Statistical analysis approaches evaluations from an objective point of view; it acknowledges that there are undefined or immeasurable variables in play. This is not a statement of impossibility; the fact is, tools have not yet been generated to quantify certain attributes. It is very clear to me, as a company commander in the U.S. Army, that morale, esprit de corps and teamwork are essential to mission success. Those terms subjectively measure the mood or attitude of a unit. Instead of ignoring “the secret” completely, we have tried to implement a quantitative evaluation with the caveat that capturing 50 percent of its significance is better than ignoring it completely. Everything considered, we developed the following equation: Speed + accuracy + consistency + competency + chemistry = gunner proficiency We understand that no equation will encompass all that is required of an M1A2 SEP gunner. The equation will have shortcomings, pitfalls and possible errors. However, the equation was developed with the attitude that an assessment, even if slightly flawed, gives the company commander a better evaluation tool than what currently exists. Speed. We defined speed as target acquisition and kill time. TA is the ability of the gunner to scan and identify targets. KT is defined by the time in between target identification and destruction. For example, if it took the gunner eight seconds to identify the target and 20 seconds to destroy the target, KT is 12 sec-SPeed Target acquisition Rank Kill time Rank Gunner 66 8.632 2 10.31154 1 Gunner 65 10.286 6 12.10348 7 Gunner 11 11.452 12 13.54423 11 Gunner 12 10.976 10 11.08654 5 Gunner 13 10.484 8 19.73269 14 Gunner 14 8.912 4 10.59423 2 Gunner 21 11.1375 11 13.70192 12 Gunner 22 9.225 5 12.512 8 Gunner 23 10.356 7 10.60577 3 Gunner 24 12.224 14 16.06154 13 Gunner 31 8.736 3 12.85769 10 Gunner 32 12.019 13 11.8734 6 Gunner 33 7.284 1 12.63462 9 Gunner 34 10.847 9 10.9873 4 Figure 1. Mean values and rankings for TA and KT. April-June 2012 5

ACCuRACy Azimuth Rank elevation Rank Gunner 66 -0.00673 2 -0.03192 5 Gunner 65 0.09746 9 -0.07428 10 Gunner 11 -0.09373 8 -0.00118 1 Gunner 12 -0.12423 12 -0.07596 11 Gunner 13 -0.0175 3 0.108542 13 Gunner 14 -0.11 10 0.007885 3 Gunner 21 0.113469 11 0.046939 8 Gunner 22 -0.1974 14 0.0112 4 Gunner 23 0.122308 13 -0.00788 2 Gunner 24 -0.0414 5 0.0522 9 Gunner 31 0.0248 4 0.1176 14 Gunner 32 0.06972 7 0.09174 12 Gunner 33 -0.0327 1 -0.01077 7 Gunner 34 0.04659 6 -0.04239 6 Figure 2. Mean azimuth/elevation values and rankings for accuracy. SPeed ACCuRACy Kill time Azimuth elevation Gunner 66 10.311543846 -0.00673 -0.03192 Gunner 65 12.10348 0.09746 -0.07428 Gunner 11 13.54423077 -0.09373 -0.00118 Gunner 12 11.086543846 -0.12423 -0.07596 Gunner 13 19.73269231 -0.0175 0.108542 Gunner 14 10.59423077 -0.11 0.007885 Gunner 21 13.70192308 0.113469 0.046939 Gunner 22 12.512 -0.1974 0.0112 Gunner 23 10.60576923 0.122308 -0.00788 Gunner 24 16.06154386 -0.0414 0.0522 Gunner 31 12.85769231 0.0248 0.1176 Gunner 32 11.8734 0.06972 0.09174 Gunner 33 12.63461538 -0.00327 -0.04077 Gunner 34 10.9873 0.04659 -0.04239 Figure 3. Analyzing speed and accuracy concurrently.

¾ Leadership: 1-10

• Leadership assessment (each platoon leader/platoon sergeant assessed their own gunners; the commander/ first sergeant/master gunner assessed all 14 gunners):

¾ Gunner’s competency: 1-10

¾ Gunner’s leadership: 1-10 For each variable of the equation, we rated the gunner’s performance from 1-14, with 14 being the highest. For “the secret” variable, we took the mean of the total scores and applied the following scale: Score of 10 = +14; score of 9 = +12; score of 8 = + 10; score of 7 = +8; score 6 and lower = +6 The process With the method in place, the challenge now was to collect data. In addition to the required training provided by the 1HBCT gunnery manual, we decided that the Advanced Gunnery Training System was the best tool to collect the data. In the past, crews who conducted unit conduct-of-fire trainer or AGTS exercises did nothing more than file or discard the printed worksheets. When was the last time leaders manipulated data provided by the system or used it as an evaluation tool? Most of the onds. We decided that the importance lies most in the gunner’s ability to destroy the target once he acquires the target. Accuracy. Accuracy is straightforward. We observed the point of impact on a target and measured its distance from center mass. The mean azimuth and elevation values on specific engagements gave us a good assessment of a gunner’s accuracy. Consistency. We measured the consistency or “shot group” of each gunner by determining the standard deviation of targets on similar engagements. Competency. The master gunner developed a 50-question multiple-choice test based on the 1HBCT gunnery manual (FM 3.20-21) and the 19K Skill Level 20 manual. We gave this test without allowing the gunners to study to capture their current tactical and technical competence. Chemistry. The most difficult variable to quantify, we conducted a survey among the gunner’s crew, the gunner himself and the gunner’s leadership. The following survey used a scale of 1-10:

• Crew assessment:

¾ Gunner’s competency: 1-10

¾ Gunner’s leadership: 1-10

• Self-assessment:

¾ Competency: 1-10 Gunner 66 Gunner 13 2.5 1.5 0.5 -0.5 -1.5 -2.5 -2 -4 2 4 Figure 4. Impact points of Gunner 66 and Gunner 13. 6 April-June 2012 time, completing all AGTS requirements was itself the task and the numbers were never used to assess crews. In addition to fulfilling the requirements established by higher headquarters, we selected five pre-basic live-fire exercises that best evaluated the gunners. The data The following paragraphs discuss the results and findings of each variable. Speed. The results of the speed variable are straightforward. An evaluator now has an objective assessment regarding TA and KT, and the ability to compare his gunners. More importantly, the evaluator can use the analysis to tailor specific training tasks to the individual gunner. I believe this concept is the most important lesson-learned from the whole process. A company commander can use the data and analysis to develop specific training tasks that directly target soldier weaknesses. For example, TA time can depict scanning techniques. Just from the raw data, an evaluator can focus on the scanning technique of C11 and C24 and determine if it is the cause of slow TA. It gives the evaluator (or, in our case, the AGTS’s instructor/operator) a focus when observing and training gunners. This data allows the IO to define and hone in on specific tasks and skillsets on which he wants to train. The next time C11 is in the gunner’s seat, the evaluator can specifically look at the Soldier’s scanning technique and determine if it is the cause of slow TA. There was also an interesting trend derived from KT. Let’s observe the data for C23 and C33. C23 acquires targets in 10.356 seconds and destroys the target in 10.60577 seconds after acquisition. C33 acquires targets in 7.284 (significantly faster than C23) but takes 12.63462 seconds in destroying the target. Again, this raw data allows the evaluator to train on specific tasks. An evaluator understands that C23 is quick to destroy the target once it is identified, whereas C33 can identify the target quickly but takes a longer time destroying it. Therefore, instead of approaching the training of both gunners equally, he can focus on improving C23’s TA ability (i.e., scanning technique) while focusing on C33’s ability to destroy targets once he identifies it (i.e., lasing techniques, trigger pull, etc). The ability to assess the proficiency of gunners and tailor training tasks to specific weaknesses is a critical tool a company commander can use in developing his fighting force. Accuracy. The data shows us the impact point of each destroyed target. A company commander and master gunner now have an assessment tool to see the exact point at which the gunner is pulling the trigger. The data shows that C21 and C23 tend to aim high on the target, whereas C31 tends to aim to the right of the target. If the key task of the gunner is to destroy the enemy, analyzing accuracy separate from speed does not give the whole picture. By comparing speed with accuracy, several interesting trends are evident. Common sense seems to dictate that the more time it takes a gunner to kill a target, the better his accuracy. This is the case for Gunner 11, Gunner 24 and Gunner 33. Gunner 12, Gunner 14 and Gunner 23 prove the opposite, as they will sacrifice some accuracy to kill the target faster. Then there are those individuals such as Gunner 13 and Gunner 21 who take longer to kill their targets and sacrifice accuracy at the same time. From this data, an evaluator can conclude which of his gunners need more training to improve these two vital skillsets. To give us a different method in analyzing accuracy, all gunner targets were plotted on a chart. (Note: Charts are produced by the AGTS system, but it is advisable to plot them on another program for ease of manipulation and the ability to view all engagements on a single chart.) Figure 4 shows the charts for Gunner 13 and Gunner 66. As you can see just from a glance at the chart, the accuracy of Gunner 66 is greater than Gunner 13. Like speed, commanders can use the accuracy variable to assess their gunners and tailor training to their needs. Consistency. Standard deviation depicts the “shot group” of each gunner. Regardless of whether the target is hit or not, the evaluator can now see the gunner’s consistency. An evaluator gets a glimpse of the gunner’s fundamentals: Is the gunner getting the same sight picture every time? Is trigger-pulling an issue? Experience clearly shows that a consistent and disciplined ConSISTenCy STd deV AZ Rank STd deV eL Rank Gunner 66 0.77494015 6 0.3486728 9 Gunner 65 0.72398746 4 0.3792348 10 Gunner 11 0.73016974 5 0.3260162 8 Gunner 12 0.49706332 1 0.2924559 4 Gunner 13 1.17426754 13 0.5422176 14 Gunner 14 0.79300372 7 0.2790062 2 Gunner 21 1.17857356 14 0.0469388 1 Gunner 22 0.62948573 2 0.310849 6 Gunner 23 0.86360988 10 0.3168184 7 Gunner 24 0.82613931 9 0.2814198 3 Gunner 31 1.05045032 12 0.4420895 13 Gunner 32 1.0234987 11 0.4234988 12 Gunner 33 0.70011686 3 0.2972346 5 Gunner 34 0.82349876 8 0.3918723 11 Figure 5. Mean standard deviation values and rankings. STd deV AZ STd deV eL Azimuth elevation Gunner 66 0.774940152 0.348672769 -0.00673 -0.03192 Gunner 65 0.723987456 0.3792348 0.09746 -0.07428 Gunner 11 0.730169736 0.326016239 -0.09373 -0.00118 Gunner 12 0.497063322 0.29245586 -0.12423 -0.07596 Gunner 13 1.17426754 0.542217628 -0.0175 0.108542 Gunner 14 0.793003721 0.279006195 -0.11 0.007885 Gunner 21 1.178573557 0.046938776 0.113469 0.046939 Gunner 22 0.629485731 0.310849002 -0.1974 0.0112 Gunner 23 0.863609878 0.316818353 0.122308 -0.00788 Gunner 24 0.826139312 0.281419797 -0.0414 0.0522 Gunner 31 1.050450321 0.442089473 0.0248 0.1176 Gunner 32 1.0234987 0.42349875 0.06972 0.09174 Gunner 33 0.700116856 0.297234615 -0.00327 -0.04077 Gunner 34 0.82349876 0.39187234 0.04659 -0.04239 Figure 6. Consistency and accuracy values. April-June 2012 7

CoMPeTenCy Test score Rank Gunner 66 21 2 Gunner 65 13 9 Gunner 11 12 10 Gunner 12 20 3 Gunner 13 13 9 Gunner 14 13 9 Gunner 21 32 1 Gunner 22 14 8 Gunner 23 15 7 Gunner 24 19 4 Gunner 31 18 6 Gunner 32 7 11 Gunner 33 19 5 Gunner 34 18 6 Figure 7. Competency test scores and ranks. Self competency Self leadership Gunner 66 8 8 Gunner 65 7 7 Gunner 11 10 8 Gunner 12 9 6 Gunner 13 4 6 Gunner 14 7 7 Gunner 21 9 9 Gunner 22 9 5 Gunner 23 8 8 Gunner 24 9 10 Gunner 31 10 10 Gunner 32 5 8 Gunner 33 8 7 Gunner 34 7 7 Figure 8. Gunner competency and leadership self-assessment. CheMISTRy Competency rate Score Leadership rate Score Gunner 66 8.6 10 7.733333 8 Gunner 65 7.46667 8 7.6 8 Gunner 11 7.952380952 8 6.047619048 6 Gunner 12 8.428571429 10 6.952380952 6 Gunner 13 5.238095238 6 5.80952381 6 Gunner 14 5.333333333 6 6.095238095 6 Gunner 21 7.380952381 8 8.333333333 10 Gunner 22 8.476190476 10 7.476190476 8 Gunner 23 8.333333333 10 7.476190476 8 Gunner 24 8.380952381 10 8.619047619 10 Gunner 31 8.333333333 10 8.095238095 10 Gunner 32 6.095238095 6 6.761904762 6 Gunner 33 7.238095238 8 5.904761905 6 Gunner 34 5.857142857 6 6.476190476 6 Figure 9. Chemistry values. method in laying the reticle on the target is necessary for success. Analysis along with accuracy of the consistency variable gives a more effective evaluation. The data allows us to tailor specific skillsets to gunners. Gunner 22 is a primary example. The accuracy (-.1974, .0112) is one of the weakest within the company; however, his shot consistency (.629485731, .310849) is among the best. Based on these numbers, the evaluator can specifically work on improving his accuracy, knowing that the gunner’s consistency displays sound fundamentals. The consistency variable provides additional information when evaluating and training gunner proficiency. Competency. The test served two purposes. First, it provided a self-assessment for the gunners and challenged them to become more tactically and technically competent. It was evident from the test scores (out of 50) that tactical/technical competency was not where it needed to be. I charged SSG Fermaint, the company master gunner, to develop a test based on what he thought were a gunner’s core competencies. He, along with the C13 tank commander, asked for input from company leadership and developed questions based on what they felt a gunner should know. The test scores served as a motivation tool and fostered an environment of learning. Second, it provided the master gunner an insight as to what core competencies were deficient. It gave a focus to the Sabot Academy classes conducted at the company level. The master gunner could now develop the curriculum focused on weaknesses rather than strengths. Like the previous variables, the data allows leadership to tailor training to individual Soldiers. Chemistry. Statistical analysis of chemistry proved to be the most challenging task. To implement a subjective variable into an objective assessment proved to be a difficult yet enlightening 8 April-June 2012

GunneR PRoFICIenCy RAnKInG Gunner score Rank Gunner 66 45 1 Gunner 33 45 1 Gunner 14 49 2 Gunner 22 55 3 Gunner 12 62 4 Gunner 23 67 5 Gunner 11 69 6 Gunner 34 62 7 Gunner 65 71 8 Gunner 21 76 9 Gunner 24 77 10 Gunner 13 86 13 Gunner 31 82 11 Gunner 32 84 12 Figure 10. overall ranking of gunners. study. Figure 8 gives us a glimpse into the gunner’s psyche. How does he view himself? Is he confident? Is he quietly confident? Does he think too little or too much of himself? Gathering these assessments gave the leadership further insight into how the gunner viewed himself. It is interesting to see not only how the gunner views himself but also how his subordinates and leadership assess his capabilities. Additional to the data below, the information from individual crewmembers and first-line supervisors are available for analysis. Figure 9 is the average ratings from all participants of the survey. It is interesting to see several examples of discrepancy between a gunner’s competency and leadership ability. Where certain gunners may be strong in competency, they lack leadership ability. On the other hand, there are certain gunners with high leadership rates who trail in competency. Which is more important? It would seem easy to conclude that competency is easier to train than leadership. Perhaps those gunners with high leadership ratings but low competency scores are younger gunners who possess tremendous potential. Having this data allows company leadership to better assess gunners within the fighting force. Most important variable Statistical analysis gives us more tools to evaluate and analyze variables. Concerning our gunner-proficiency equation, which variable is the most important? Answering this question is significant in that it allows us to prioritize our training. This can be applied in two ways:

• In a limited time, knowing the most significant variable allows a company to focus on the one variable that will most improve overall gunner performance.

• When training new gunners, understanding the most significant variable focuses the master gunner on the key task that will foundationally grow a proficient gunner. To process this raw data and find which, if any, variables are statistically significant, I challenged SPC Mark Rothenmeyer to use a linear-regression model.7 Linear regression creates an equation to explain a dependent variable; in this case, the gunner’s ranking among the company gunners, using one or more independent variables. In this equation, the independent variables were TA time, KT, average error of azimuth, average error of evaluation, standard deviation of azimuth, standard deviation of evaluation, score on the KT and leadership ranking. The equation is derived as: Gunner ranking = -17.24 +1.28 (TA) +0.36 (KT) +12.13

(AEA) +25.71 (AEE) -0.55 (SDA) +11.08 (SDE) +2.58 (KT)

+0.27(LR) The regression resulted in one variable being the most statistically significant: TA time. In the equation, 78 percent of the variance in the dependent variable explains this independent variable. This regression supports the hypothesis. It found that finding the target – an unrated technical skill – was statistically significant in the overall ranking of a gunner, while the gunner’s leadership rating proved to be less significant. This is not to say that a gunner’s ability to lead his Soldiers is not important, but that a gunner’s evaluation should be on more than just the traditional subjective criteria. It is clear from the LR model that if a company has limited time to improve the performance of all its gunners, it should focus on TA. Furthermore, the model shows that when training brand-new gunners, TA is the most significant variable. Conclusion The purpose of this study is to use lessons-learned from the statistical-analysis revolution in sports to better train and evaluate Soldiers. I argue that using objective assessments provides a better evaluation tool to measure the proficiency of Soldiers and information to tailor training according to their weaknesses. Furthermore, it provides a company commander and master gunner the tools necessary to best place and use Soldiers while creating an environment of competition and esprit de corps that drives and motivates Soldiers. Not all Soldiers are equal. They have their individual strengths and weaknesses. Statistical analysis provides a definition to that statement. On which tactical or technical individual Soldier task do they need the most training? With statistical data, I can develop a gunnery-training plan with my master gunner that uniquely targets each Soldier’s weakness. Specificity is the key here. The more information I have, the better I can create training plans. When a gunner now enters the AGTS, the master gunner or IO now has specific information that can enhance that training experience. Figure 10 allows the company commander and first sergeant to better deploy and use gunners. Company leadership can use this information to place proficient gunners in key positions. For example, with the high rankings of Gunner 33 and Gunner 22, they are now prime candidates to become a commander, executive officer, platoon leader or platoon sergeant gunner. A company commander now has an objective assessment of how his gunners compare to each other. It is not the whole story, and there are variables that are not quantified (or correctly taken into consideration), but it is more informative than what he previously had. That is the key: any information that can improve the commander’s assessment of his Soldiers improves the fighting force. A company commander can now assess who his seventh or 10th best gunner is, and that information will pay dividends in future operations. Competition and esprit de corps are key components in war fighting units. How does a leader maximize the competitive spirit within individual Soldiers while building camaraderie? Although not a priority when starting this study, the byproduct April-June 2012 9

Acronym Quick-Scan AeA – average error of azimuth Aee – average error of evaluation AGTS – Advanced Gunnery Training System FM – field manual hBCT – heavy brigade combat team Io – instructor/operator KT – kill time of competition among gunners proved to be valuable for the unit. It humbled and motivated gunners to continue to develop in their profession. The desire to outperform peers created an environment that encouraged gunners to pursue excellence. The company fostered an environment that encouraged time spent on gaining knowledge and acquiring skill. It is important for leaders to understand that a very thin line exists between healthy and toxic competition. Company leadership should implement competition in an environment that maximizes its potential. Effective leadership is a multi-faceted challenge. Statistical analysis will not provide all the answers and may serve to be only a small fraction of the equation. However, I believe it is a tool that can directly enhance a commander’s capability of successively training and deploying a killer fighting force.

CPT Michael Kim commands Company C, Task Force 1-72 Armor Regiment, at Camp Casey, Korea. He has served in various positions in the United States and Iraq, including scout and support-platoon leader, 1st Squadron, 1st Cavalry Regiment, Budingen, Germany; executive officer, Alpha and Headquarters Company, 1-35 Armor Regiment, Baumholder, Germany; and commander, Forward Support Company, Operation Iraqi Freedom 08-09. He holds a bachelor’s of science degree from the U.S. Military Academy at West Point. SPC Mark Rothenmeyer serves as an M1A2 SEP tank crewmember in Company C, Task Force 1-72 Armor Regiment, Camp Casey. He has also served as training room noncommissioned officer in charge, loader and driver with Company C, 1-72 Armor Regiment, Camp Casey. He holds a bachelor’s of science degree in economics from Frostburg State University. notes 1 Bestsellers list July 6, 2003, New York Times, http://www.nytimes. c o m / 2 0 0 3 / 0 7 / 0 6 / b o o k s / b e s t - s e l l e r s - j u l y - 6 - 2 0 0 3 . html?pagewanted=all&src=pm. 2 Grabiner, David, “The Sabermetric Manifesto,” baseball1.com, January 2004, http://baseball1.com/baseball-archive/sabermetrics/sabermetric-manifesto/. 3 Ibid. 4 Investigation result on the sinking of the RoK ship Cheonan, RoK Ministry of National Defense, May 20, 2010, http://www.mnd.go.kr/webmod-ule/htsboard/template/read/engbdread.jsp?typeID=16&boardid=88&se qno=871&c=TITLE&t=&pagenum=3&tableName=ENGBASIC&pc=und efined&dc=&wc=&lu=&vu=&iu=&du=&st=. 5 Kim, Hyung-Jin, and Kim, Kwang-Tae, “Korea Attack: Yeonpyeong Island Shelled by North Korea,” Huffington Post, Nov. 23, 2010, http:// www.huffingtonpost.com/2010/11/23/korea-attack-yeonpyeong-island_n_ 787294.html#s189509. 6 Simmons, Bill, The Book of Basketball, New York: Ballantine Books, 2009. 7 A key strength of the U.S. military is the collection of unique backgrounds and talents consolidated in one fighting force. SPC Mark Rothenmeyer was an economic major in college and decided to join the military to serve his country. I was able to use his unique ability to assist me in this study. LR – leadership ranking oPneT – operator new-equipment training RoK – Republic of Korea SdA – standard deviation of azimuth Sde – standard deviation of evaluation SeP – system-enhancement package TA – target acquisition 10 April-June 2012 When visiting Fort Benning, please stop in at any of our library locations:

‰ Armor Research Library Bldg. 5205 (harmony Church)

‰ donovan Research Library Bldg. 70 (Main Post) For 24/7 access, visit http://www.benning.army.mil/ library/ …the places where Soldiers go for information 24/7 wherever they are in the world. MCoe Libraries