Game Theory in Management

Since it looks as though this will be my last blog for February, and ProjectManagement.com’s theme for February is creativity, I thought I’d use the cover of “creativity” to address a topic that, I’m confident, has gone unaddressed by hundreds of PM-centric writers over decades of publications, with articles and blogs running into the tens of thousands: Did the Munchkins from The Wizard of Oz do Project Management? If so, were they any good at it? Does an evaluation of Munchkin PM offer up some insights to us on this side of Oz?

I think we can safely assume that the Munchkins did do Project Management. How do I know? Check the YouTube® clips from when Dorothy arrives in Oz, specifically near the main square of the Munchkins’ village. There are several buildings (presumably houses, but none of them have signs or features that would indicate one way or the other), an impressive fountain, an extremely flat cobblestone pavement, and, of course, the Yellow Brick Road. In addition, the visible buildings appear to be in an architectural style consistent with what one would find in an 18^th century English town in the countryside that has been hit with an excess charm bomb (in the megaton range), with a seizure-inducing color-enhancement feature. If the Munchkins were ducking in and out of caves, that would be one thing. But, given that their village is presented the way it is, it’s inescapable: those civil engineering artifacts didn’t create themselves. The Munchkins clearly “did” Project Management.

Were they any good at it? I think that, in approaching this question, we must be careful about our basis of comparison. GTIM Nation members know that among my favorite axioms is “All things fail by irrelevant comparisons.” We have no way of knowing what the Munchkins’ cost or schedule baselines were, so we’re left with comparing their buildings and infrastructure to the other examples within The Wizard of Oz. These include:

• The Emerald City. This would probably be an unfair comparison, given that the Emerald City is presented as the zenith of architecture and building expertise in all of Oz, and the Munchkins’ village is just a place in the countryside.
• The Wicked Witch’s Castle. A foreboding place, it is, nevertheless, an impressive piece of construction. Garrisoned by what appears to be around twenty soldiers, it looks like it could withstand a prolonged assault by thousands. We’re not told if the Wicked Witch of the West had it built herself, or if she inherited it, or how she came to be in possession of it, but I think it’s a safe bet that the winning contractor for this project wasn’t headquartered in the Munchkins’ village, which leaves us with…
• …Dorothy’s house. Granted it’s somewhat worse for wear for having been scooped up by a tornado and flung who-knows-how-far; but, having been dropped on the Wicked Witch of the East while she was in the town square of the Munchkins’ village, it does provide a good opportunity for comparative architecture and building technique analysis. The house itself is a simple plank-on-frame affair, made entirely of wood. It is in remarkable shape for having been scooped up by a tornado and dropped who-knows-how-far, but that’s kind of the point. The Munchkins’ houses/buildings look like they are made of polished stone, which would tend to make them impervious to random cyclones running over them.

Based on the only valid comparison left to us, that of comparing the level of building expertise on display between the Munchkins’ village and Dorothy’s house, one would have to conclude advantage: Munchkins, meaning that they not only “did” Project Management, they did it well, at least in comparison to Kansas-based construction project contractors of the same vintage.

Which leads us to the last of my posed questions, is there anything we can learn from Munchkin PM? Before legions of GTIM Nation wise-guys enter comments to the effect “Michael, you know that those Munchkin buildings you’re going on and on about are really just painted backdrops, right?”, I want to focus on things that didn’t happen when Dorothy arrived on the project site in Munchkin Land. My thinking is that, had the Munchkins retained an “advanced” capability in one of the PM disciplines that is mistaken for insightful PM, say, risk management (veteran GTIM Nation members totally saw this coming), then the scene immediately after Dorothy emerges from the dropped house would have unfolded in a dramatically different fashion. Consider that no risk analyst in the universe can possibly restrain himself from pointing out that, had they documented a specific “risk event,” with odds of occurrence and impact to the cost, scope, and schedule baselines thoroughly analyzed, and that event were to actually take place, that they actually predicted as such. It’s simply not in their capacity from engaging in the rarely-encountered-but-completely- documented “I told you so.”

But the Wicked Witch of the East had clearly been antagonizing the Munchkins for some time, since they broke out into an obviously rehearsed song of joy on the occasion of her being squished by Dorothy’s house. It follows that, had any Munchkin risk manager performed an analysis of any of their considerable public works projects, that one of the identified risks would have to be “Wicked Witch of the East arrives, antagonizes the Project Team, leading them to flee the job site, leading to X impact on schedule.” And, if such a risk analysis did actually exist, with the recommended work-around being “Wait for a house from a completely different place to drop on her head,” then I would become the risk managers’ biggest fan. Interestingly, none of that happened.

As it is, I think we can conclude the following:

• The Munchkins did do Project Management,
• They were good at it, and…
• Even they knew that risk management was a waste of time.

As I’ve discussed in previous blogs, much of the Project Management codex describes techniques and strategies that have been tested many times and found to be effective in circumstances both common and rare. Of course, many decisions facing PMs are not specifically addressed in this codex – otherwise, a robot (perhaps resembling the B-9 robot from the original Lost in Space television series) could be programmed with the PMBOK Guide®, (we would all be out of a job) with verbiage including:

• “Placing Organizational Breakdown Structure elements in a Work Breakdown Structure does not compute.”
• “By rejecting Earned Value analysis techniques, you are a quack.”
• “Warning, warning! Spending budget on a risk analysis is counterproductive!”
• …and the ultimate, “Danger, Will Robinson! The accountants are attempting to generate the cost performance reports based exclusively on data from the general ledger!”

Fortunately for we PM practitioners, the actual performance of the PM function can only be reduced to a series of canned strategies on a highly abstract and uneven basis – hence the need for creativity (ProjectManagement.com’s theme for February), often in large doses, in order to bring in our projects on-time, on-budget, on a consistent basis.

I would like to take a moment to look at “creativity” from the perspective of a PMBOK Guide® --fluent robot. In Game Theory there are whole series of canned strategies available for use in various scenarios, or games, viewed as analogous to real-life situations. Whenever the situations varied significantly from their recommended strategies, though, the practice of alternating the use of different canned techniques is known as a “mixed strategy.”

Referring back to that bedrock of Game Theory, the Nash Equilibrium, the use of mixed strategies makes computing the equilibrium very difficult. Since it’s been awhile since I’ve defined it, here’s one definition, from Dictionary.com:

(in economics and game theory) a stable state of a system involving the interaction of different participants, in which no participant can gain by a unilateral change of strategy if the strategies of the others remain unchanged.[i]

Ah, but that’s the rub with mixed strategies, isn’t it? If the PM were to find herself in a project that presented as a “stable state,” (highly unlikely, but let’s continue with this), but did not adhere to one given strategy, then Thomas Nash himself proved (mathematically) that there would be at least one Nash Equilibrium. More specifically, “Nash proved that if we allow mixed strategies (where a pure strategy is chosen at random, subject to some fixed probability), then every game with a finite number of players in which each player can choose from finitely many pure strategies has at least one Nash equilibrium.”[ii] The other players could not know for certain beforehand if changing their existing strategies would result in a superior payoff, or not. Now let’s throw in the fact that, with extremely rare exceptions, no projects will present in stable states. Even the most cookie-cutter of efforts, performed over and over, will be subject to forces and circumstances that all but eliminate the very notion of a completely repeatable environment. Also consider that the strategies of not just the members of the Project Team, but members of the organization outside the Project Team, members of competitors’ organizations, and just plain random players who become involved for a multitude of reasons ranging from geographic proximity to owning shares in vendors’ companies – all of these people’s strategies come in to play, and cannot possibly be quantified, much less compared on an objective basis. And make no mistake – as Metcalf’s Law (a.k.a. “The Butterfly Effect”) makes clear, very small perturbations in some (or even one) nodes of a complex network can have a cascading, cataclysmic impact on a large number of nodes on far-flung aspects of that network. In other words, that one, lowly person who’s involved in your project simply due to geographic proximity? If he happens to be the chairperson of the Neighborhood Association that brushes up against your project’s job site, things could get dicey, and fast.

What should become clear from such an analysis is that creativity is arguably the essential element of Project Management. I’m not discounting the value of the codex of PM research – far from it. Instead, I’m arguing that we, as PM practitioners, have a very long way to go before we can program our robots to fill in for us.

Unless, of course, like the B9 robot from Lost in Space, there’s actually a person inside of it, and that person is very creative.

[i] Retrieved from https://www.google.com/search?client=firefox-b-1-d&q=nash+equilibrium on February 15, 2020, 21:52 MST.

[ii] Wikipedia contributors. (2020, February 9). Nash equilibrium. In Wikipedia, The Free Encyclopedia. Retrieved 16:03, February 17, 2020, from https://en.wikipedia.org/w/index.php?title=Nash_equilibrium&oldid=939985159

It continues to be a good 2020 for our friends, the risk managers. While they still haven’t earned initial caps status when I refer to them, I am continuing the trend from the last couple of weeks’ blogs and averting my poison-pixel gaze, and towards another one of my favorite targets, albeit one that I don’t pick on too often. I’m talking about another set of friends, the quality managers (who have been granted initial cap status previously, but no more). Why are quality managers deserving of GTIM Nation scrutiny? Well, it all starts with their pronouncements.

Recall my oft-repeated axiom of management, that of Quality, Availability, Affordability: pick any two. Anybody with any significant amount of management under their belts will recognize this as being absolutely true, and really not worth debating, like the assertion that pro wrestling in the United States is far more staged theater than competitive sporting event. Ah, but not our friends, the quality managers. I have yet to meet one that isn’t convinced that the tactics, procedures, and management approaches germane to quality management should absolutely be embraced and executed, the sooner the better, cost is irrelevant, by every organization and project team in existence, and to fail to do so is akin to chargeable management malpractice. This attitude becomes all the more remarkable when one considers how the American National Standards Institute (ANSI) develops standards on various management science topics. When PMI® asked me to participate in the ANSI Standard that they were then developing for Earned Value Management Systems, this participation included a trip to Pennsylvania to attend various workshops to help train authors to properly contribute to an ANSI Standard. Aside from the tutorials involving copyright infringement and general writing style, an actual representative from ANSI hosted a workshop on content. I’ll never forget one of this fellow’s guidance recommendations: we were told to write in such a way that no one who could be considered an “expert” in the field we were writing about could disagree with the verbiage we were providing.

It was at that point that I began to doubt that the eventual product would actually advance the practice of Earned Value Management. Why? Because in order to get the document to a point where no one would disagree with it, it would have to be content-reduced to such an anodyne state that nothing even coming close to insightful would pass the no-expert-disagreement test. It’s as if no one even considered that this race to the lowest common denominator would naturally steer content in the opposite direction of genuine quality. With such parameters in place, creativity (ProjectManagement.com’s theme for February), ironically enough, didn’t have a place in the document-creation cycle. This isn’t the way to generate quality anything. Quite the opposite, it virtually guarantees that any management science-themed standard attaining the ANSI approval imprimatur would be extremely weak sauce, promising to not offend any palate exposed to it. So, I asked myself, if this approach extended to ANSI documents about quality management itself, would they be as insubstantial? Well, I checked it out, and they did not disappoint.

The document I’m referring to is entitled “Quality Management Principles,” by the International Organization for Standardization (ISO), located in Geneva, Switzerland.[i] This is, by any professional standard, a poor-quality document. It’s essentially 20 pages of a priori assertions, sentences phrased in the weak passive voice, and sometimes a priori assertions phrased in the weak passive voice. I do not know if ISO operates under the same “no expert should disagree” theme that ANSI appears to follow, but I do know from my days as an advertising agency apprentice what a “posinon” is. Short for “positive, inferential non-statement,” a posinon is a rather irksome aspect of many an advertising campaign. “Coke is it!” is a classic example.[ii] They don’t really assert anything that can be evaluated as True or False – they just sort of sound good, or insightful. Examples of posinons in the text of “Quality Management Principles” include:

1. “To manage an organization effectively and efficiently, it is important to involve all people at all levels and to respect them as individuals.”[iii]
2. “Enable self-evaluation of performance against personal objectives.”[iv]
3. “Manage risks that can affect outputs of the processes and overall outcomes of the quality management system.”[v]

Let’s take a look at these one by one, shall we?

1. This is clearly absurd. Does one really need to “engage” the HVAC technicians, and “respect them as individuals,” for, say, an investment brokerage firm on Wall Street to be managed “effectively and efficiently?”
2. If my personal objective is to get rich young and retire early, how does “enable(ing) (my) self-evaluation” help the organization advance its quality objectives?
3. Okay, I didn’t want to go here, and stated as such at the beginning of the blog; but it really is impossible to “manage risks that can affect outputs,” since almost anything can literally affect outputs, including such unmanageable things as the weather, chances of key operating personnel contracting influenza, and availability of goods or services from vendors trying to deal with the weather and key personnel contracting influenza.

Also, just out of curiosity, can any of these assertions be tested empirically? I mean, if an organization is successful, could it point to having “involve(d) all people at all levels and …respect(ed) them as individuals?” And, if the same organization performed poorly on a project, what evidence could be brought to show that they had failed to do such “involving?”

I could go on (and often do), but you see my point. Such posinon-premised a priori nonsense has no place in any management science codex, much less Project Management. So, my creative suggestion for improving the quality of Project Management would be to reject the quality management guys’ “guidance” until such a time that they return to real, testable premises as the basis for their arguments and conclusions.

Or they could simply claim “quality management is it!”

[i] Retrieved from https://www.iso.org/files/live/sites/isoorg/files/store/en/PUB100080.pdf on 8 February 2020, 21:16 MST.

[ii] Retrieved from https://www.coca-colacompany.com/news/history-of-coca-cola-advertising-slogans on February 10, 2020, 18:59 MST.

[iii] Retrieved from https://www.iso.org/files/live/sites/isoorg/files/store/en/PUB100080.pdf on 8 February 2020, 21:16 MST.

[iv] Ibid, pp. 7

[v] Ibid, pp. 9.

The central problem when discussing creativity in the Project Management realm (ProjectManagement.com’s theme for February) is that we’re looking at two very different sides of the same coin. On the one hand, you have the templates, procedures, and canned strategies germane to the PM universe, including Work Breakdown Structures (WBSs), Earned Value and Critical Path Methodologies (EVM and CPM, respectively), Scope Management, and all the rest of the codex in the PMBOK Guide®. As anyone who has endured the PMP® examination process can attest, these concepts are all rather formalized and structured. However, the other hand is a very different schema indeed, where it’s common knowledge among PM practitioners that the ability to bring in a project on-time, on-budget is heavily reliant on finding innovative solutions to the unexpected problems that arrive with a bang at the PM’s front door, solutions that rarely have a standardized, template response. “How shall we find the concord of this discord?” (A Midsummer Night’s Dream). I think that the central question to finding the balance between what’s regarded as standardized techniques in approaching Project Management problems and having the management latitude to do whatever the heck the nominal PM wants to do hinges on one question: Does it work?

Not to put too fine a point on this, but I would like to point out the two criteria for evaluating whether or not something can be considered scientific, even in Management Science space:

• It has to be observable. Explaining phenomena that no one can perceive is as useless as arriving at an 80% confidence interval on a risk analysis.
• It has to be repeatable in an experimental setting. Theories that “explain” one-off occurrences are always suspect.

Based on this criterion, establishing the WBS early in the project is scientifically sound. This technique has been shown to serve well as the basis for the subsequent cost and schedule baselines time and again, while project work bereft of the same is almost always given to failure for any but the smallest, simplest of projects. Likewise, both Earned Value and Critical Path analysis are pretty standard for all but the most basic projects – if you want to have a fighting chance of bringing them in on-time, on-budget, that is. Which brings us to addressing our dichotomy: when is adopting a creative management strategy, nominally at odds with the standardized approach to managing a project well, a good idea? As GTIM Nation knows, any Game Theorist worthy of the name (and initial caps) will resort to a payoff grid, so:

In Scenario 1, the “Doesn’t care enough” bin doesn’t mean that the subject PM is visibly lackadaisical. It very well could be that they are simply so entrenched in their own approaches to PM problems that they can’t be convinced to abandon said approaches. When those canned strategies actually work, then these may be successful in spite of a lack of consideration of alternate, creative tactics.

The PM who cares enough to use tools known to work, but keeps an open mind when either they themselves or a member of the Project Team suggests a novel solution or approach, is going to be highly successful. This scenario is closest to describing The Gamesman archetype from Michael Maccoby’s brilliant work The Gamesman: The New Corporate Leaders (Simon and Schuster, 1976), which, based on Maccoby’s research, consistently experiences the most success.

Now, about that “don’t be dumb” mention in the blog’s title: if the named PM isn’t sufficiently motivated to do what it takes to bring in the project’s scope on-time, on-budget, one way of avoiding the use of the PM tools known to enhance success would be to listen to the Earned Value and Critical Path naysayers, and point to a “creative” management approach that eschews these tools. Experienced Project Controls Analysts have seen this scenario play out time and again, and, yes, they think those PMs trying to pull it off are dumb. Really dumb.

Finally, if the named PM really wants to perform well on the project, but only listens to our friends the Accountants and risk managers on how to do so, then they’re not really doing Project Management at all, but pursuing an obviously deficient derivative.

And I don’t do obviously deficient derivatives in this blog.

As promised at the end of last week’s blog, I’ve finally found a purpose for risk management, though I still refuse to do initial caps when referring to it (for alert GTIM Nation members who would say “You just used initial caps in the title of this blog for “risk management,” my response is, I use initial caps for each word in my titles). Before I reveal this quirky and yet valid reason, let’s take a quick view of the epistemology of Management Information Systems, or MISs (note the initial caps).

The key valid Project Management information streams have the following things in common:

1. They provide the actual decision-makers the data they need to make informed choices on their projects,
2. They further support the creation of an audit trail, or the (truthful) narrative on how the project has been executed, and the circumstances behind that execution,
3. …and they’re deserving of initial caps (e.g., Earned Value Management System, or Critical Path Methodology).

As I have previously pointed out in this blog, risk management utterly fails in number one, and therefore doesn’t qualify for number three. How do I know this? Because of the very definition of risk management offered up by some of their leading figures.

In a point-counterpoint article I contributed to in PMNetwork magazine many years ago, my risk management counterpart defined its purview as encompassing anything that might happen to a project, positive or negative, the latter being described as traditional risk management, the former belonging to “opportunity management.” Of course, management time is finite: the management information system that doesn’t accurately differentiate between where the project is in trouble and where it’s doing fine, and thereby pointing out where PM attention is needed is, well, useless. By their own admission, risk managers maintain that their specialty’s purview covers all things, positive or negative, needing of attention or not.

I’ve addressed many of the other reasons that risk management is such a disappointment in the delivering-relevant-information department in previous blogs, but a quick recap includes:

• Risk analysis (MSWord forces the first letter in a bulleted list to have initial caps) reports tell the PM exactly what? That there’s an X% chance that something will go wrong, resulting in $Y in added costs, or Z days in delays? Is this helpful?
• Especially when one considers that each of the parameters in the preceding bullet is little more than a wild-donkey guess.
• Then there’s their insufferable tendency to couch everything in Gaussian curve terminology (Carl Friedrich Gauss was a person, so I had to use an initial cap there).

“Alright, already, we know your objections to risk management” I can hear the two or three risk manager members of GTIM Nation (note the initial caps) saying, “but you promised a valid use for us! What is it, already?” I’m glad you guys asked.

From the second numbered entry above, Project Management MISs (at this point I’m basically trolling the risk managers with the whole initial caps business) provide a type of audit trail. I would further divide this audit trail into two bins:

• It generates the historical narrative, useful for future PMs to learn how best to tackle similar projects, and
• It gives customers who have, shall we say, an exceptionally developed curiosity about the choices being made by the Project Team, their basis, sensibility, etc., a sense that they’re in the middle of the day-to-day decisions that execute the project.

It’s this second bullet where risk management may actually be indispensable. Think about it: if your customer is hovering over your project so closely that you have to process a Baseline Change Proposal just to go to the restroom, there’s no better way of getting them to ease up a bit than by showing them that you have not only considered all of their worries, but you have “quantified” them in Gaussian curve jargon, and run a series of statistical analyses that would induce granite statues’ eyes to glaze over.

“If everyone minded their business,” the Duchess said in a hoarse growl, “the world would go round a great deal faster than it does.” (Alice in Wonderland, 1865).  So, here are the risk managers, making the world go round a bit faster, and for this they deserve recognition, even in the GTIM blog. I would like to point out, though, that had the risk management community reached out to me and explained the real purpose behind their schtick being to thwart overly hoverish customers (think about it: what customer is going to needle the PM who’s just waiting for an opportunity to discuss the significance of the gap between the median and average time for a given task?), then much of the pixel ink I’ve spent making fun of them could have been saved.

On second thought, and upon reviewing some of the risk management hate e-mail I’ve received, nevermind analyzing what might have been. The odds wouldn’t be good that they would ever receive initial capitalization status.