Game Theory in Management

“The secret is to work less as individuals and more as a team. As a coach, I play not my eleven best, but my best eleven.”  -- Knute Rockne

It seems to me that much of the discussions on managing talent (ProjectManagement.com’s theme for February) revolves around a simple binary, illustrated in the Game Theorist’s favorite tool, the Payoff Grid, below:

When attracting or retaining “talent,” in those instances where you perceive a member of your Project Team to be talented, and they are, in fact, capable, then it’s all okay. Similarly, if someone wants to be part of your Team, and you perceive that they really can’t contribute, and this person, indeed, would add little or nothing to your Team, then the PM can deal with them on that basis. There’s a great deal of pixel ink spilled on the inherent injustice of Scenario 1.A., where a genuinely talented person has been judged to be unlikely to contribute, and how horrible such misjudgments can be, blah blah blah. I will touch on that shortly, just not in a conventional manner (GTIM Nation expects no less). For now, though, let’s look at Scenario 2.B, where a person has been hired on or transferred/assigned to your Project Team based on reputation, or education, or other factors that would normally point to a “talented” person when that person is not a serious contributor and may, in fact, ultimately detract from your Team’s cohesion and overall effectiveness.

The Impact Of Filter Conditions On Success

Based on the Payoff Grid above, the central question that needs to be resolved is: How to bridge the gap between what the PM perceives as “talent,” and who is actually “talented?” In the instance of Scenario 2.B., the person who has been perceived as talented when they actually aren’t pose a severe danger to your project’s success. These people didn’t attain the trappings of being highly capable without any true merit by accident. In these cases, it is entirely likely that the personnel who present as more capable than they really are belong to the Maccoby Archetype “Jungle Fighter,” who gets ahead via political machinations and skullduggery more than any genuine achievement. If members of this archetype have infiltrated your Project Team, they can (and will) cause major headaches, and can easily ruin your chances of coming in on-time, on-budget.

Let’s go back and re-evaluate the criterion used for assessing “talent.” If those criteria are valid, then it would take a great deal of effort for the not-talented person to join your Project Team. If those criteria are invalid, it’s much easier, as illustrated by another payoff grid:

Scenarios 4.A and 4.B represent the dangers involved when the PM is attempting to attract talent, hence the title of this particular blog. Note that, based on that particular row in the Payoff Grid (#4), what the two danger scenarios have in common is that the hiring manager is using either invalid criteria in an otherwise effective candidate-assessment process, or, in extreme cases, the whole assessment process is invalid. Attracting and retaining a highly-capable Project Team is tough enough as it is; but, if the selection criteria are flawed from the get-go, and sub-capable personnel are added to your Team while talented ones join only through extreme luck, guess what’s gonna happen to your overall capability. Keep in mind, also, that there’s an elevated possibility that these people who presented as capable while not really being so are Maccoby Archetype Jungle Fighters, meaning that overall Team cohesion is on a collision course with some major stress tests.

Circling back to the scenario where a truly talented individual is passed over due to bad selection criteria in the assessment model, what happens to those people? You’re not using them, and they don’t just evaporate. It’s reasonable to assume that they will find a position somewhere, and it’s entirely possible that that somewhere is going to be with your competition, either inside the macro-organization or with its outside competitors. This particular management nightmare scenario, where you’ve made the effort to attract talent but ended up rejecting it, has become a proverbial double-whammy: your ability to accomplish scope on-time, on-budget has been diminished, while your competitor’s has been enhanced.

Want to avoid that scenario? Make it a point to continually re-evaluate the criteria you use to add (or dismiss) members of your Project Team, to test for validity and relevance. Do this successfully, and who knows? You may get quoted in some ProjectManagement.com blog 90 years from now!

GTIM Nation knows of my respect for Michael Maccoby, particularly his book The Gamesman: The New Corporate Leaders (Simon and Schuster, 1976). In it, Maccoby posits four basic archetypes of workers in the corporate world:

• The Gamesman sees his particular industry as a type of game. Because of this, he is more likely to have mastered the “rules,” as well as being more likely to take chances.
• The Company Man tends to assume the persona of the team around him.
• The Craftsman doesn’t care so much about who he works for, but cares a great deal about the quality of his particular output.
• The Jungle Fighter gets ahead through political machinations and deceit more than actual performance.

If Maccoby is correct (and I believe he is), then the savvy PM would gain an organizational behavior and performance insight by correctly identifying which of the archetypes most closely match the members of the Project Team. The problem with placing your team members into bins, though, is that if you make a mistake, the damage to Team morale could easily exceed the benefit of optimal task assignments. What’s the savvy PM to do?

I’m a firm believer in the idea that elements of a person’s speech, from range of vocabulary to tempo of delivery, can reveal much about that person’s level of education, background, and, most importantly, thought processes. The way a person writes can be even more revealing. And, as luck would have it, PMs have access to a regular source of the Project Team’s (or, at least the Work Package and Cost Account Managers’) writings, the Variance Analysis Report, or VAR. For the sake of this exercise, we’ll assume that a particular Cost Account has encountered out-of-threshold negative cost and schedule variances. If your Team’s real-life VARs contain elements of the example styles below, you just might have a reliable identification method for the Maccoby archetypes!

The Craftsman:

Variance Analysis: Both the negative cost and schedule variances are due to the amount of rework required to get the scope right on this Control Account. Whomever agreed to the original scope baseline should have taken into account the fact that the novel technical approach, using largely untested technology, might result in our inability to achieve the reliability and quality standards set out in the Technical Scope Documents.

Corrective Action: We shouldn’t have ever agreed to use an untried technical approach in the first place. But, now that we’re here, we should immediately issue a stop work order to the subcontractor, and complete the work using the traditional engineering approach. We may carry this variance to completion.

The Gamesman:

Variance Analysis: We were aware that the subcontractor’s use of this new technology carried an element of risk with it, which is why this Work Package is referenced in the risk management (no initial caps) plan. It should be pointed out that, had this new technology performed as promised, this Work Package alone would have saved more money than the entire fixed fee amount.

Corrective Action: Simply file a Contingency Baseline Change Request, and reference the entry in the risk management (no initial caps) plan as justification.

The Company Man:

Variance Analysis: Technically, this variance should not have happened. My entire Work Package Team has followed the company’s procedures to the letter, and I have personally surveilled time sheet entries to guarantee that no one has mischarged down to the nanosecond.

Corrective Action: Procedure KHG-10008874-A3Cd, subparagraph E, recommends that, in the event of an out-of-threshold negative cost variance, where the offending WP is referenced in the risk management (no initial caps) plan, that a Contingency BCP be prepared. However, Procedure DIK-877640-H8B4, subparagraph N, states that, simply because an event makes an appearance in a risk register doesn’t mean that it should automatically be considered a change in scope. I will need to consult with company executives to see how they think I should handle this variance.

The Jungle Fighter:

Variance Analysis: Perhaps if some of the members of the Proposal Team – whom I won’t name, out of professional courtesy – were to show more fealty to this company’s standards and stated goals, this variance wouldn’t have happened in the first place. If it weren’t for my friends Ken and Jerry over in Risk Management (of course Jungle Fighters think that the term “risk management” deserves initial caps) we would be in grave difficulty.

Corrective Action: We should prepare and submit a Contingency BCP. If it gets hung up in the Baseline Change Control Board, we could make an emergency appeal to the BCCB’s Deputy Chief, who has a very close relationship with the customer’s representative, if you know what I mean.

Now, this listing of writing styles should in no way be considered definitive. But if you recognize the utility of attracting Gamesmen and Craftsmen, using Company Men where they can contribute best (in secondary or support roles), and getting the Jungle Fighters off of the Project Team, then these stylistic nuances might be your best indicator of those archetypes.

Also, I’ll thank GTIM Nation to refrain from doing exactly what I prescribe in this blog, and identifying my Maccoby archetype based on my writings down in the comments section. I don’t want to risk my reputation of complying with ProjectManagement.com’s directions, after all. If you know what I mean.

I’m old enough to remember when risk management (no initial caps) first wormed made its way into mainstream PM guidance. One of its funniest manifestations was an assertion that, if a given project could assert that it was really, really not risky, then many of the conventional Cost/Schedule performance measurement systems should then be considered superfluous, and therefore not mandated for government work. Of course, non-risky work didn’t need all of that high falutin’ risk analysis performed, at least when it came to making the case for the imposed level of rigor of the Project Management Information Systems (PMISs) applied to the work. Essentially, the early forays of risk management (no initial caps) into the PM world worked to undermine its overall utility, a vehicle for sending those pesky Critical Path and Earned Value zealots packing.

Flash forward to the introduction of Agile/Scrum Project Management. I believe that one of the main reasons that Agile/Scrum received such quick and broad acceptance in the Information Technology (IT) PM world was due to the fact that it addressed a widespread problem in that arena, that conventional configuration management/ change control techniques were inadequate for software engineering scope. A typical Baseline Change Control Board, made up of senior project and customer managers, would meet only once per month. Then, if a Baseline Change Proposal/Request (BCP/BCR) wasn’t considered acceptable by every member of the BCCB, it would be back to the proverbial drawing board to address the concerns of the Board, and a re-submission, to be considered next month. Agile/Scrum represented a way around such an ossified process. The people who were in a position to approve relatively minor baseline changes – the kind that happen all the time in IT projects – could do so in near real-time, and the new scope could be pursued in a far timelier manner, all without inviting the stigma of having a rubber baseline.

What’s a “rubber baseline?” It’s the bugaboo of us traditional PMs, the equivalent of this line from Shakespeare’s Henry VI, Part I:

Is this the Talbot, so much feared abroad
That with his name the mothers still their babes?

To have a rubber baseline means that you can change cost, or schedule, or (by implication) scope informally, almost at will. Variances magically go away, and, with them, the integrity of your project’s cost/schedule measurement systems. To the generation of PMs who still think that quoting Shakespeare is hip, to unknowingly engage a rubber baseline is to self-identify as completely inept, and to do so knowingly is to “smile, and smile, and be a villain.” (Hamlet, Act I, sc. V).

Interestingly, even though Agile/Scrum was originally developed for Information Technology projects, many other sectors have adapted it, including manufacturing. Even so, this approach may not have delivered on its highly-touted potential. According to one source, 75% of IT executives believe that their projects are “doomed from the start.”[i] Naturally, one has to wonder what the proximate cause of this sense of inevitable doom could be – is it hardware? Lack of talented personnel? Sub-standard coffee delivery service? Or could it be … the Project Management approach?

Since both Agile/Scrum and the implementation of a rubber baseline have in common a weakening (or even elimination) of the Configuration Management/Baseline Change Control process, what could we say separates, or distinguishes them? Let’s use the Game Theorists’ favorite tool, the payoff grid:

Members of GTIM Nation will quickly observe that the only okay scenario here is 1A. All of the others represent trouble, either because the project is being run by someone who doesn’t have the awareness that he shouldn’t be using a rubber baseline and, therefore, is likely to crash the project based on sheer ineptitude, or because he’s a villain who has successfully circumvented the very cost/schedule performance measurement systems that would alert his superiors when the project is in trouble. Don’t misunderstand – I’m not arguing against the use of Agile/Scrum in all cases. The brutal truth of the matter is that traditional approaches to maintaining baseline integrity had become moribund for projects being executed in a fast-moving environment, like Information Technology, and some change had to come on this front in order to keep PM relevant in those business situations.

But to remove all of the guardrails of traditional Configuration Management/Change Control, in the name of granting the PM more latitude of action, simply invites a management environment where “you’ll never find a more wretched hive of scum and villainy.” (Yeah, that’s from Star Wars. Were you expecting more Shakespeare?)

[i] Retrieved from https://blog.capterra.com/surprising-project-management-statistics/ on January 12, 2022, 16:47 MST.

The canned strategies that Work Package or Control Account Managers (WPM/CAM) use when determining how to handle any float associated with their project activities in a Critical Path Methodology (CPM) network are highly analogous to how most people deal with their New Year’s resolutions, proving once again that PM techniques are far more closely associated with real life than their Asset Management counterparts.

Take, for example, the timing of when we begin to pursue our New Year’s resolutions. Many of my past resolutions dealt with diet or exercise regimens, so I’ll use those as comparison points to how WPGs/CAMs manage float. For those PMs who don’t deal with Critical Path Methodology schedule networks on a regular basis, “float” is the amount of time non-critical activities have in excess of their originally estimated durations before a late finish can be expected to cause the entire project to finish late. There are several ways of managing float in an activity, but many parameters go in to informing the best way of approaching it.

PMs who tend to want to begin their activities at the earliest possible opportunity (meaning all predecessor activities are complete, and the resources for the activity are available), should consider the following:

• If the actual output of the task is a tangible object, and it gets finished early, will it need to be stored somewhere in-between the time it’s ready, and the time when that activity’s successor is ready to receive it?
• If the answer to the previous question is “yes,” is there a facility or area where this thing can be stored, without cost or penalty? If your project is to add to the population of a zoo, and the acquisition of the tigers activity has some float, you might want to make sure that that procurement activity’s successor, finish the tiger enclosure, is 100% done prior to completing this activity. Putting them in with the great apes would be a really bad idea.
• Those resources you are tapping – are they also needed in other activities happening at the same time? Few things get uglier in Project Portfolio reviews than the conflicts over resource availability, but one of those things occurs when the other CAMs realize that you are, in fact, in a position to wait a bit before laying claim to said resources.
• Are there any references to your activity in the risk register (no initial caps)? Not that it matters, I just don’t want to give the risk managers any “I told you so” moments should your activity go long, with a “risk event” having been pinned to it.

Now, let’s contrast these strategies to how we manage our New Year’s resolutions.

• Did you resolve to actually create an object? If you begin too soon, and finish before, say, Labor Day, your spouse may be led to believe that your objective was too easily attained. Best to wait for a while before beginning!
• If the answer to the previous question is “yes,” what does that portend for expected follow-on projects? For example, if the object that you made was an archway trellis for your back yard, will the aforementioned spouse also expect a flower garden, or water fountain?  Best to wait until the growing season is over prior to even starting the arch!
• Your resolutions no doubt require your time and energy. What of the other things that require your time and energy? Why are you spending time and energy at the gym when there are parties to host and attend? Caution with your resource utilization decisions can benefit overall resolution portfolio performance significantly (you can quote this line verbatim to sound like you are truly managing your resolutions).

In those instances where you may be falling behind in attaining your New Year’s resolutions, another PM-themed remedy is available. Whenever a WBS element at the reporting level has an out-of-threshold negative cost or schedule variance, the analyses provided in the Variance Analysis Reports almost never go straight to “poor performance.” Instead, the following are usually cited:

• Poor original estimates in the baseline(s).
• Subcontractor unavailability.
• Material/Equipment/Parts rate changes.
• Poor or sub-standard conditions (e.g., weather for construction projects).

Did you resolve to lose weight or exercise to get in better shape, but aren’t advancing as you had planned? Consider offering the following explanations:

• “I didn’t realize that losing one pound per week was considered unnecessarily aggressive.”
• “The gym I wanted to use isn’t open when I’m available to work out.”
• “The whole COVID thing has made workout equipment, especially free weights, prohibitively expensive.”
• “I was going out for a jog this morning, but the temperature is 25 degrees!”

Conversely, if these management approaches to handling your New Year’s resolutions strike you as being inferior to, say, calculating the Return on Investment (ROI) for them, you’re an Asset Manager, aren’t you?

Waaaaayyyyy before PMI® became a professional association in 1969[i], many large-scale, complex projects were being successfully completed on-time, on-budget, by managers who had an extremely advanced grasp of the field of PM, even if they didn’t articulate their understanding in terms we use today. I could be wrong about this – any day now the Construction Office chamber of the Great Pyramid could be discovered and accessed, with the schedulers still waiting for their early-version Critical Path Methodology software to complete a forward and backward pass on their 20,000-activity network. But my speculation is that these historic PMs would do things like assign a higher priority to time-critical tasks without invoking the term “crashing the schedule.” I’m also fairly certain that, prior to 1823 and Carl Friedrich Gauss’ publication of the monogram Theoria combinationis observationum erroribus minimis obnoxiae, PMs were quite aware of many of the things that could go wrong with their projects, they just didn’t document them with some statistical speculations of their odds of occurring in a “risk register” (no initial caps here, either).

Once PMI® did come into existence, and much of its approach to the management sciences took on a scholastic flavor, the lexicon not only became standardized, it became a bit more academic, a function of applying its wide range of theories across multiple industries. PMs working the American Interstate Highway system in 1958 may have had little in common with those working in the nascent National Aeronautics and Space Administration (NASA), but managers in both organizations would have instinctively known that some activities would have to be completed before others could actually start, and that comparing those activities’ percent complete to their cumulative spent cost and schedule could give them a pretty good idea of how much their work would cost at completion, and how long it would take. They would have understood the basics of schedule logic and cost/schedule performance indicators, even if they didn’t use those exact terms.

And therein lies a problem: many in PMI® today have come to the practice of Project Management on scholastic or theoretical terms, while many others have arrived with a boots-on-the-ground understanding of its precepts. This latter approach is how I swerved into PM. I was working for a Department of Defense contractor that was building a type of advanced communication system. The design and development of this system came with a myriad of research deliverables and design reviews that were described in a document known as the Contract Deliverables Requirements List, or CDRL. Trick was, these deliverables’ due dates weren’t firmly established. They were almost always described as being 90 or 180 days after some other deliverable was submitted, or 30 days prior to one of the design reviews. My title at the time was “Data Manager,” and it was my job to “coordinate” the development of all of these deliverable documents, oversee their progress, and transmit them to their far-flung recipients on-schedule. At the time personal computers were something of a novelty, but I had one in my office, running one of the earliest spreadsheet packages, Lotus 1-2-3. So, I went through the CDRL, one entry at a time, and in the cells I had labeled “Due Date,” would place the equation to add 90 (or whatever) days to that deliverable’s predecessor’s end date. Without having been taught what a finish-to-start link was, or the term “lag,” I ended up constructing a Critical Path network that would automatically re-calculate a myriad of Start and Finish Dates based on the values that were placed into the Today’s Date field, and whichever Review date was considered reliable.  It wasn’t until the Project Controls Analyst on this project saw what I was doing, and told me that I should look in to the Project Management profession that I had any idea that that’s what I had been doing all along.

So, back to the communications gap. Having spent a lot of time around construction and manufacturing PMs, as well as their Agile/Scrum counterparts, I realize they’re discussing the same ideas, just using a different lexicon. So, as a service to GTIM Nation members who may find themselves in a situation where they have to serve as a translator between the Practical and Scholastic-oriented PMs, I offer the following conversion table.

This is far from an exhaustive list, but it will suffice for now.

Or, there’s more, but I’m done.

[i] https://www.pmi.org/about/learn-about-pmi/founders