Game Theory in Management

Although I generally like to avoid examples in extremis, I would like to review a virtual pandemic that hit a virtual world, one that I discuss at length in my second book, the book this blog is named after, Game Theory in Management. The subtitle to that book is the same as the subtitle to this blog, too: modeling business decisions and their consequences. The virtual plague took place in the World of Warcraft game, which is typically referred to as an MMORPG, or Massive Multiplayer Online Role-Playing Game. Its “world” is Azeroth, and players belong to one of two warring populations. However, within these populations are sub-groups, or tribes if you will, who do not necessarily work together in combatting their nominal enemies.

Into this setting was introduced a disease, the “Corrupted Blood” incident. In the words of Joel Hruska, from ExtremeTech, fifteen years ago,

The boss of the Zul’Gurub raid instance, Hakkar the Soulflayer, had a debuff he could apply to nearby players that caused damage every few seconds. The debuff was designed to kill players quickly enough that anyone without healer support would die fairly quickly.

Unfortunately, Blizzard made a mistake. Hunter pets, if put away while the debuff was applied, would still have it when they were pulled back out again — like, say, in a populated area. That was the first problem. The second problem was that Corrupted Blood was infectious, and spread to people nearby. The third problem? NPCs (non-person characters) could catch it. When they did, they didn’t die. They just transmitted it to everyone within range, indefinitely.[i]

In my research while writing Game Theory in Management, I came across stories of those characters with healing powers offering their services at inflated prices (supply and demand hold sway over virtual worlds as well as our own, it would seem), characters deliberately teleporting infected animals into enemy’s population centers (biological warfare), and, perhaps one of the more chilling virtual behaviors, characters would assume the role of journalists, approach population centers that may or may not have been infected and therefore deserving of quarantine, assess the situation from afar, and then re-position themselves at nearby road junctions to inform passerby of that center’s status – or not. It seems some characters were known to, shall we say, adjust the message depending on whether or not the recipient was of their own tribe or clan.

I would like to point out that, in spite of entreaties from Blizzard Entertainment for the players of World of Warcraft to initiate behaviors that would eradicate the plague within the game itself, insufficient numbers of players did so, necessitating a hard re-set of the entire shebang in order to rid Azeroth of the disease. In short, when called on to behave a certain way as to benefit the entire population, individual players in sufficient numbers elected to act out on their own interests, which prevented any but the most drastic of solutions from actually working.

Meanwhile, Back In The Project Management World…

It has been my experience that the go-to implementation strategy for advancing Project Management within an organization (particularly large ones) entails some form of the following steps:

1. Tap into the energy generated by the organization’s realization that an advancement in PM is needed, usually due to some project (or projects) going off the rails in dramatic fashion.
2. Hire or promote “experts,” and create a Project Management Office (PMO), or at least a team that does PM.
3. The PMO/Team performs an analysis of the level of PM maturity in place, and creates a document for advancement.
4. The PMO/Team continues to generate documents, this round creating official policies or procedures, that mandates certain PM-related practices and artifacts.
5. Some projects will embrace the new techniques, but many will seek to opt out, either by asserting a basis for exemption from the procedures, or by engaging the silent veto/slow roll tactic.
6. To the extent that the organization’s major projects have improved their PM practices, overruns and late project completions are reduced, and the amount of energy that went into the creation of the PMO begins to dissipate.
7. With no major overruns or delays to justify the expense of all those experts in the PMO on the payroll, and with significant amounts of project work not adhering to the PM procedures, the need for the PMO begins to be challenged in the organization’s executive suites, until…
8. Another project disaster happens, and the cycle begins anew.

The behaviors of (most of) the organizations in this approach mirrors those of Corrupted Blood-inflicted Azeroth, in that no amount of urging individuals to act for the common good but not necessarily for their own benefit will bring about the desired macro-organizational change. Self-interest will always manifest itself – not universally, to be sure, but in sufficient amounts to keep the desired organization-wide change/maturity advancement from occurring. It’s simply human nature. That being the case, the solution to the self-interest problem must include…

Look at that! Out of pixel ink for this week. Tune in next week for tips on advancing PM in a macro organization driven by individuals predominantly looking out for themselves.

And please stay safe in this world.

[i] Retrieved from https://www.extremetech.com/gaming/307717-researchers-wow-corrupted-blood-plague-to-understand-coronavirus-infections on April 5, 2020, at 19:58 MDT.

Yes, you read this blog’s title correctly. The cliché, of course, is “the best and the brightest,” as if, upon representing these two populations in a Venn Diagram, the two circles would perfectly overlay each other. But, as I wrap up Game Theory In Management’s take on ProjectManagement.com’s March theme of leading PM trends, I’d like to examine those tests that organizations tend to employ when screening for sufficiently talented personnel to join their project teams, and their implications for advancing Project Management theory writ large.

Probably one of the most common tests is the number and type of college degrees held by the candidate, combined with their Grade Point Average, or GPA. In the hard sciences, these criteria come close to exclusivity. Make no mistake, I’m not necessarily against this standard. For example, without consistently high grades and a very good showing on the MCAT, students aren’t even admitted to medical schools, and I’m completely okay with that. But I would like to point out that, in business within a free market economy, the GPA element might be overrated. Steve Jobs and Bill Gates, arguably the most successful business executives ever, actually dropped out of college. And in an article for CNBC.com by Kathryn Dill on Eric Barker’s book Barking Up The Wrong Tree,

A survey of over 700 American millionaires found that their average college GPA was 2.9. “College grades,” Barker writes, “aren’t any more predictive of subsequent life success than rolling dice.”[i]

If one considers the highly dynamic world of management, this probably won’t come as that much of a surprise. After all, what’s included in attaining high grades in school? In addition to hard work and some level of talent, a large dose of conformity is needed for such success. Although never articulated out loud, it’s a safe bet that all successful students are fully aware of the folly of correcting any teacher on any topic, even if (or especially if) said teacher goes off on a political or social-economic rant while supposedly teaching an utterly unrelated subject, like mathematics. GTIM Nation is well aware of my opinion of some of the foundations of current college-level management courses, predicated as they are on the absurd notion that the point of all management is to “maximize shareholder wealth.” In essence those who do really well in college-level business courses are taught two things that are utterly incompatible with Project Management, specifically:

• The Asset Managers’ take on business is the only correct one, and
• Really good managers will employ the staid analysis methods stemming from the previous assertion when they get a job in the real world.

Don’t misunderstand – I’m completely aware of the fact that the newly-assigned PM who has a working understanding of Work Breakdown Structures, Earned Value and Critical Path Methodologies, Cost Performance Reports in Format 1, Gantt Charts, and all the rest will have an immense advantage over her less-educated rivals. But throughout my time in Graduate Management School, which included two semesters of Accounting, two semesters of Finance, one of Statistics, and the usual Organizational Behavior and Performance and Strategic Management stuff, only once in the entire curriculum did any professor mention any of the skills from the previous sentence, and that was a one-class exercise in developing and analyzing a schedule network (in an Information Technology class, no less).

If GPA is a potentially flawed metric to use in predicting future performance, …wait! What did I just write? Did I use the phrase “predicting future performance?” Yes, yes I did. This calls for … the risk managers! Given the problem of assessing which Project Team candidates will represent the type of talent needed to advance PM in general, or come up with trend-setting ideas in the field in particular, what data should we feed our Monte Carlo or Decision-Tree analysis? That’s a really tough question, since, in the United States at least, it’s basically illegal for large organizations to consider any of the following factors when hiring:

• Gender
• Race
• Religion
• Marital Status
• Age
• Social-Economic status

…among others, and I’ve just established that, for PMs anyway, use of where the candidate went to college and the GPA is at least suspect. In short, for this predictive function, there’s simply no data that could be fed into two of the risk managers’ favorite tools that could possibly result in a usable information point.

Sooo…where does that leave us in our quest to find those who are most likely to do some PM trend-setting? Invoking Sir Arthur Helps is where that leaves us. He was supposedly the first person to write “Nothing succeeds like success.”[ii] The true trend-setters in PM are not those writing out their a priori assumptions ad nauseum, nor are they to be found in (most) college settings. They will be the ones who actually bring their various types of projects in on-time, on-budget, usually unceremoniously, but also usually consistently.

These might not be your brightest, but they’re your best.

[i] Retrieved from https://www.cnbc.com/2017/05/24/what-happened-to-your-class-valedictorian-probably-not-much.html on March 29, 2020, 18:05 MDT.

[ii] Helps, Sir Arthur, Realmah, 1868, retrieved from https://www.phrases.org.uk/meanings/261100.html on March 30, 2020, 17:55 MDT.

GTIM Nation knows of my fondness for Shakespeare. In The Merchant of Venice, the lady protagonist, Portia, is to be wed the suitor who correctly chooses one of three caskets that contains her portrait. The caskets are made of gold, silver, and lead, and bear the following inscriptions:

• Gold: Who chooseth me shall gain what many men desire.
• Silver: Who chooseth me shall get as much as he deserves.
• Lead: Who chooseth me must give and hazard all he hath.

The male protagonist, Bassanio, chooses the SPOILER ALERT! STOP READING NOW IF YOU DON’T WANT TO KNOW HOW THIS PART OF THE PLAY WORKS OUT! lead casket and, with the dowry that comes with marrying Portia, is positioned to help rescue his friend and financier, Antonio.

Meanwhile, Back In The Project Management World…

In previous blogs I’ve discussed an axiom that’s undeniably true, but seems to receive very little notice in management science discussions. PM’s own version of the three caskets: that, for any given good or service,

• Available, high quality is going to be expensive,
• Available, affordable is going to be of lower quality, and
• Affordable, high-quality is going to require a wait.

Proceeding from this premise, and circling back to ProjectManagement.com’s theme for March, of leading PM trends, I have to wonder which of the three characteristics is going to be satisfied or improved by the next major technological breakthrough in PM. Whichever technological advancement riddle is being addressed will have significant implications, including those ideological opponents who can be expected to be arrayed against the innovator. Recall also previous blog discussions on Thomas Kuhn’s profound work The Structure of Scientific Revolutions, that the maintainers of the current thinking have a vested interest in diminishing (if not out-and-out thwarting) the overturning of those theories that undergird their own “expertise” (or Portia’s hand in marriage, whatever).

For example, consider some future PM researcher who discovers a way to reliably pull actual (or a reasonably accurate) percent complete figures for all active Work Packages. That’s right: no Project Controls Specialist pestering the Control Account Managers in the last days of the reporting period for their percent complete estimates, only to be told “this is a level of effort task” when it clearly isn’t, and entering the 8.333% more-done-than-last-month silliness into the cost processor. Without the missing or clearly biased percent complete estimates, this new solution would reliably feed the organization’s Earned Value Management System (EVMS) with accurate Estimates at Completion (EACs, both cost and duration), Performance Indices (both cost and duration), current and cumulative variances (both … well, you know) – all quickly and easily. A quantum leap in quality in Management Information for the same (or even less) expense would have been attained, easily summoning the following opponents:

• Our friends, the accountants. They’ve been taught (ad nauseum) that all relevant cost information originates in the general ledger, and generally predict at-completion costs by assessing spending trends. And yet, here’s those pesky PM-types running laps around their very own Management Information System track. The value between bookkeeper and accountant just got much narrower, and none of the latter category will be happy about it.
• Some of the people supposedly within our own camp, specifically the risk analysts (no initial caps) and the estimators. The risk analysts want to use Gaussian curves to predict the future of performance on a project, but EVMSs can do so automatically, with a ten-point accuracy bracket, which the risk analysts could only dream of. Similarly, the estimators are given to developing an EAC by re-forecasting the remaining scope, and adding that figure to the project’s cumulative actuals. Not only would our automated and quality-controlled EVMS perform that function automatically (Budgeted Cost of Work Remaining + ACWPcum), but it can do one better: simply calculate the EAC. So, no, neither the risk analysts nor the estimators would be happy with this innovation.
• Don’t get me started on the guidance-generating organizations that waste spend so much pixel-ink on the need to “properly” match the percent-complete method to specific types of scope.

Similar opposition can be safely assumed to come about if the next innovation solves one of the other two riddles, Affordability or Availability. A system that’s both affordable and readily available will draw fire for being of low quality, even if the value of the MIS stream being made available performs at least as well as others deemed to be worthy – I’ve encountered it time and again. The only question then becomes How long will the innovation to make PM more affordable, available, or of higher quality be opposed before its benefits overcome its automatic objections?

Once such objections are made, and found to be wanting, those articulating the criticisms would do well to recall the words of the Prince of Morocco, after learning that Portia’s portrait is not in the gold casket he selected: “Portia, adieu. I have too grieved a heart To take a tedious leave: thus losers part.” (Merchant of Venice, Act II, Scene VII)

Shorter version: I picked wrong, I’m outta here.

ProjectManagement.com’s theme for March, Leading PM Trends, is conversely both broad and focused, germane to what we do while also maintaining that elusive characteristic of existing in the future (which, again, cannot be quantified). Along these lines I would like to explore with the rest of GTIM Nation another duality that I’ve witnessed over and over again in organized efforts at advancing PM capability within an organization, and it has to do with an axiom from our near cousins, the Information Technology PMs. This axiom may be among the most tightly-packed six words in all of the management science codex, and it is this: begin with the end in mind. So, here’s my question: what end do we have in mind?

I think that, generally speaking, there are two main trains of thought here. One camp seeks to change the behavior of the targeted organization, to lead, compel, or force them to “do” Project Management, and do it to the level of robustness these self-appointed experts are expecting. The other camp seeks to place into the hands of decision-makers the information they need to make, well, informed decisions. Experienced members of GTIM Nation already know where I stand; but, for newcomers and for those who need a little more than a priori assertions to justify the acceptance or rejection of management science theories, I’m going to do a little exploring.

The “change behavior” crowd, it seems to me, dominate the common thinking here. Their exhortations in conference paper presentations, scholarly articles, and boardroom laments are as predictable as they are tiresome. These people simply know that their organizations would be much, much better managed if the principles would adopt their principals, set up a Project Management Office, and encode their insights into must-obey company policies and procedures. If possible, these ones become even more insufferable when they attain their PMP® Certifications, as if the entire weight of PMI’s academic progress was behind their unworkable implementation strategies. Their end, or targeted end-state? Of course they want to see all project work (and maybe a little bit of the work that clearly is not project-related) captured in Work Packages, which roll up to Control Accounts via Work Breakdown Structures, and feed data to Earned Value and Critical Path Management (EVM and CPM) systems, variances captured and explained in Variance Analysis Reports, consistent with the thresholds described in the Project Management Plan, Change Control Boards meeting regularly to review well-written Baseline Change Proposals, etc., etc. But it goes beyond that: they also want to witness Control Account managers eagerly awaiting their Cost Performance Reports (in Format I, of course) and Gantt Charts, clutching them to their hearts as they grab their hard hats and rush out to the job site, suddenly illuminated as to the best course of action to bring their projects in on-time, on-budget. But, given the choice between managers who don’t care at all about these artifacts of doing “proper” PM who, nonetheless, bring their projects in on-time and on-budget on a consistent basis, versus those who do the CPR (Format I) and Gantt-Chart-clutching, but somehow manage to finish late and over-budget, the change-the-behavior crowd would actually prefer the latter.

On the other had we have those who simply want to place into the hands of the decision-makers the information they need to maximize the odds of bringing their projects in on-time, on-budget. This is actually far simpler than the change-their-behavior crowd would ever admit, but it is demonstrably true. Want to know how much your Work Package or Control Account will cost at its completion? Divide the cumulative percent complete into the cumulative actual costs. Yes, it’s that simple. The same is true for duration. Divide the percent complete into the total number of days since it started, and you have its most likely duration, accurate to within ten points. For those whose purpose is to put into the hands of the decision-makers the information they need, such techniques are simply golden. Bereft of such superfluous requirements as tapping a master resource dictionary, or (GTIM Nation totally saw this coming) risk analysis, the project controllers become highly effective, but also extremely efficient. And they become so with hardly any changes to the way the actual project teams actually perform their work – sometimes, with no changes at all. This makes them infinitely more popular with the personnel performing the scope than their change-behavior counterparts, who are far more likely to disrupt the in-place business models, and on a massive scale.

Here’s the ironic part of the whole duality: whereas the change-behavior crowd virtually always fails in attaining their desired end-state (due to silent veto or slow-roll counter-strategies employed by the project team), the provide-the-information team almost always succeeds in changing the organization’s behavior. Even with the most rudimentary cost/schedule performance information being bandied about at the most casual of project review meeting venues, those previously given to ignoring (or even rejecting outright) PM precepts will quickly find themselves discussing negative cost variances being offset by positive schedule variances, and processing change orders to mitigate scope creep.

So, let’s circle back to the question underlying how best to advance PM capabilities within the organization: advance to where? Are we changing behavior, or generating key information streams as innocuously as possible? These are two very different destinations. It would be prudent to know where the PMO is headed.

Do you members of GTIM Nation know very many rich PMs? I know it kind of depends on one’s definition of “rich,” but the only truly wealthy PMs I know just happen to own companies that provide project controls specialists to contractors who run large very large facilities. That’s not to say that former Project Managers are rarely among the ranks of upper echelon of executives – quite the contrary, virtually all of the highly-placed executives I’ve met have at least some experience as a PM. A natural take-away from these anecdotal data points is that PMs become relatively wealthy once they stop doing Project Management exclusively, and bring that level of expertise to a different position, one that requires a broader-base of management expertise than can be attained through Project Management alone.

My thinking is that such a state of affairs is messed up. There simply has to be a way for the more advanced Project Managers to amass more wealth than, say, our friends the accountants (to be fair, the only wealthy accountants I know own their own bookkeeping or tax firms). We do, after all, have a far more advanced and nuanced grasp of management science writ large than those following the simplistic “maximize shareholder wealth” mantra. But the accountants have a very powerful ally in the management world: the taxman. In what has to be one of the both earliest and most powerful examples of a political/governmental structure essentially ensuring the success of one specific variety of business discipline, it’s simply illegal to run a commercial enterprise without doing accounting. One has to employ the double-entry bookkeeping system, and do so following a very specific (and massive) set of rules, or else risk imprisonment. Why? Because governments are funded by the taxes these companies generate, and the amount they generate is based on their books, which are set up and maintained by our friends, the accountants. So, running your business based on the information stream emanating from the accountants’ data feed seems natural, enlightened even. What’s a Project Manager to do?

Let’s start by stipulating that the eat-your-peas-style hectoring has never worked, does not currently work, and will never work. One of my oft-stated axioms is that you cannot advance a capability maturity by leveraging organizational power (you can’t make people get better at a thing). A readily-discernible corollary is that you can’t nag them into improving, either. Oh, sure, harping about how managers who eschew Earned Value or Critical Path methodologies are really dummies who don’t deserve their executive positions might provide some level of catharsis to the presenters at certain PM-centric conferences, but I seriously doubt they change anybody’s mind. And, if you think about it, the presenters at these conferences, with very few exceptions, actually pay money to push their ideas on the attendees, which is the exact opposite of monetizing PM. So, of the potential methods for monetizing PM, let’s scratch “nag executives into recognizing the value of the presenters’ cliched positions, and pay them and their like-minded colleagues more” off the list, shall we?

Perhaps one of the most compelling arguments for monetizing any given business strategy would be to establish its ability to predict the future. Another Hatfieldian axiom is that, if you could give even a poor stock trader a copy of next week’s Wall Street Journal, such a one could pretty much start picking out the color of his soon-to-be-purchased Jaguar (the high-end car make, not the predatory feline). This is where our friends, the risk managers (no initial caps) come in, since they’re in the business of using Gaussian curves to do just that: quantify the future, in a way that provides meaningful information to PMs. Of course, pointing to a high-value information stream and actually delivering one are two very different things. And when I say they are in the business of providing such a feed, I really mean they make their particular take on management science appear attractive through a combination of two strategies: asserting that they can provide some sort of meaningful quantification of the future (they can’t), and the aforementioned eat-your-peas approach, but with a snarky overlay that shifts the narrative to “anyone not doing risk management is doing a poor job of managing in general, and Project Management in particular.”  Risk management (they only got one initial cap because the term started the sentence) is a big business around the world, so I suppose this combination of pseudo-management science pushes is somewhat successful in monetizing PM. However, astrology is also big business, so there’s that.

Then there’s probably the most appropriate method of monetizing PM, that of documenting the clear link between those managers who have an appreciation and working knowledge of it, and their measurable instances of out-performing their PM-devoid colleagues. This gets rather tricky, though, due to that facet of human nature illustrated by the axiom “Victory has many fathers, but defeat is an orphan.” Failed managers will never admit to having crashed their projects due to a lack of competence, while successful projects will have a seemingly never-ending list of business model proponents, anxious to pin the success on the presence of their favorite PM-related theme.

I don’t know, maybe the best way of monetizing PM is by making snarky comments about its detractors in a weekly blog, and awaiting a soon-to-be-realized payday…