Institutional theory (Meyer & Rowan, Reference Meyer and Rowan1977) holds that a firm must reflect legitimacy, signaled by the extent to which it resembles other firms. Without it, a firm’s value may appear limited, since stakeholders perceive the firm as irrelevant or undesirable (Suchman, Reference Suchman1995). Value, however, requires the presence and communication of uniqueness (MacDonald & Ryall, Reference MacDonald and Ryall2004; Makadok & Ross, Reference Makadok and Ross2013) by offering products and services perceived to be different from those of their rivals (Peteraf & Barney, Reference Peteraf and Barney2003). In short, to survive and develop a competitive advantage, firms must appear legitimate enough but also be sufficiently different from other firms.
Striking a balance between novelty (i.e., uniqueness) and risk poses critical tensions for decision-makers. While multiple techniques have been suggested for ensuring legitimacy (Castelló & Lozano, Reference Castelló and Lozano2011; Lamin & Zaheer, Reference Lamin and Zaheer2011; Richards, Zellweger & Gond, Reference Richards, Zellweger and Gond2017), uniqueness is limited because most rivals follow the same rules. Indeed, imitating the conventional norms may seem safe and legitimate, but it erodes competitive advantage (Adner & Zemsky, Reference Adner and Zemsky2006). On the contrary, entirely avoiding the conventional norms violates what is known of risk assessment and long-term value creation (Cai, Shi & Jiang, Reference Cai, Shi and Jiang2025). Importantly then, to warrant such deviations, successful strategists must understand the long-term benefits of deviating from conventional norms. The extent to which those deviations prove beneficial, however, is often unclear.
To clarify this issue, we examine how minor adaptations to an otherwise legitimate and conventional strategy can outperform that same conventional strategy, despite the added risk. Using theoretically supported heuristics, we deploy both approaches on the decisions participants could have made in a fee-based sports pool with weekly and seasonal prizes. Using nine years of real results from the games on which those decisions were applied, we found distinct differences in outcomes when applying the strategies. One approach consistently chose favored teams, while the other deviated from that strategy in a small way, purposefully and systematically choosing certain underdogs. Those underdogs were the least likely to be selected, offering more novelty to individual decisions. Our results suggest that adopting the second approach, which blends convention with nuance, yields greater long-term profitability. Indeed, choosing seemingly risky selections can be less risky, in terms of overall performance, than strictly following conventional norms. Throughout the paper, we discuss the contest details, our approach, our findings, and their strategic implications.
Literature review
Competitive strategy involves choices and trade-offs. In other words, firms succeed by pursuing opportunities that align best with organizational resources and capabilities, providing the greatest potential payoffs (Berchicci, Dowell & King, Reference Berchicci, Dowell and King2012; Parnell & Crandall, Reference Parnell and Crandall2021; Vinther Larsen & Gulddahl Rasmussen, Reference Vinther Larsen and Gulddahl Rasmussen2018; Wu, He, Duan & O’Regan, Reference Wu, He, Duan and O’Regan2012; Zhou, Lu & Chang, Reference Zhou, Lu and Chang2016). Circumventing or minimizing the trade-offs is challenging. Each strategic alternative has projected returns and risk, but identifying the best option is not always straightforward (Brenes, Montoya & Ciravegna, Reference Brenes, Montoya and Ciravegna2014; Dess & Davis, Reference Dess and Davis1984; Lee, Hoehn-Weiss & Karim, Reference Lee, Hoehn-Weiss and Karim2021; Parnell & Wright, Reference Parnell and Wright1993). This problem is fundamental to strategic decision-making, but the proliferation of the internet and artificial intelligence adds another layer of complexity as choices and opinions become more visible to potential competitors (Leppänen, George & Alexy, Reference Leppänen, George and Alexy2023; Sabaruddin, MacBryde & D’Ippiloto, Reference Sabaruddin, MacBryde and D’Ippiloto2022). To address these problems, strategists seek to pursue opportunities that offer promising benefits while minimizing potential drawbacks. More commonly, choices that limit downsides also dilute the potential for upsides. Moreover, it is imprudent to either pursue too much risk or overlook risky opportunities. Since strategic risk is unavoidable, decision-makers must try to manage and mitigate it (Michel, Reference Michel2023).
Firms manage strategic risk in different ways. They can hedge it by purchasing a form of insurance (Shiu & Yang, Reference Shiu and Yang2017). Other options include contractual agreements that transfer risk to suppliers, partners, or insurance firms, and working capital or other financial buffers are less necessary for those risk-sharing agreements (Teece, Peteraf & Leih, Reference Teece, Peteraf and Leih2016). Evaluating how risk influences strategic choices is complex because measuring it is difficult, especially when a strategy is novel or the payoffs from resource or capability combinations are unclear (Genus & Coles, Reference Genus and Coles2006; Noy & Ellis, Reference Noy and Ellis2003; Soltanizadeh, Abdul Rasid, Mottaghi Golshan & Wan Ismail, Reference Soltanizadeh, Abdul Rasid, Mottaghi Golshan and Wan Ismail2016). Hence, managers often fail to properly evaluate the risk associated with a strategy or develop alternative approaches for mitigating it. Some firms leverage stakeholders to combine complementary resources or capabilities, thereby reducing transaction costs and risk (Dyer & Singh, Reference Dyer and Singh1998; Palmatier, Dant, Grewal & Evans, Reference Palmatier, Dant, Grewal and Evans2006). Doing so can support mutual goals, promote greater relationship continuity, and facilitate positive word-of-mouth communication, loyalty, and knowledge exchange. In this respect, firms manage risk by co-creating value (Prahalad & Ramaswamy, Reference Prahalad and Ramaswamy2004). Nonetheless, this approach seeks to manage risk by sharing potential gains and losses, and limits risk more than it enhances performance. That is, limiting risk draws more focus than supporting opportunities, and the trade-off hinders performance.
Indeed, strategic choices involve trade-offs (Haffar & Searcy, Reference Haffar and Searcy2015). Choosing one goal means aligning resources to support the selected tactics and trading the possible results that may have been realized had other goals been pursued instead. For strategists, achieving both economic and non-economic goals involves establishing some organizational uniqueness. Said differently, for organizations to thrive, they must be able to showcase their unique characteristics effectively. On the one hand, organizational efforts must mimic other organizations; otherwise, external assessments of legitimacy fall short (Meyer & Rowan, Reference Meyer and Rowan1977; Risi, Vigneau, Bohn & Wickert, Reference Risi, Vigneau, Bohn and Wickert2023). While these efforts may appear quite different, they must reflect the norms of society (Boso et al., Reference Boso, Amankwah-Amoah, Essuman, Olabode, Bruce, Hultman and Adeola2023; Suchman, Reference Suchman1995). Aims to reflect legitimacy exist to proactively and defensively counter potential assessments of illegitimacy. For offensive efforts, organizations go beyond coping with their environment (Leppänen et al., Reference Leppänen, George and Alexy2023; Porter, Reference Porter1979) and devise efforts to provide offerings that buyers would prefer, even at higher prices (Makadok & Ross, Reference Makadok and Ross2013). Indeed, the tension between being legitimate enough and also being different enough remains problematic. Organizationally, assessing when deviation from accepted norms might generate unique value presents an important and intriguing question: how far should a firm deviate from legitimacy to achieve uniqueness?
At a minimum, answering this question would clarify trade-offs inherent in the struggle for legitimacy and uniqueness. To that end, we provide an example to illuminate how and where strategists can enhance potential performance. First, we demonstrate how choices can retain legitimacy by adhering to industry norms, and more importantly, how to deviate from those norms by making novel choices that do not unnecessarily increase risk. Second, we show where those choices could be made. In the following narrative, we outline the parameters for our exploratory approach to answering our research question and the data used to apply our decision rationales, relate the data setting to business strategy, explain the theoretical support for our generic rationales, and discuss the results of our investigation. Had our rationales been used in these competitions over the nine National Football League (NFL) seasons, the approach appearing most risky (unique) would have shown more reliable, stable profits than its more mainstream, legitimate counterpart.
Theoretical framework
Organizations’ need for legitimacy is rooted in institutional theory (Meyer & Rowan, Reference Meyer and Rowan1977). But too much focus on following competitors impedes firms’ ability to morph into organizations capable of meeting evolving environmental needs (DiMaggio & Powell, Reference DiMaggio and Powell1983). Given the limited empirical evidence from the business world to support the investigation of our research question, we analyze example data from a growing industry to explore how combining novelty with traditional strategies might improve outcomes. Although present in society for centuries, sports betting has experienced rapid growth since the U.S. Supreme Court overturned the federal government’s ban on sports gambling outside of Nevada. As of 2025, it is legal in Washington, D.C., and 39 U.S. states (American Gaming Association, 2025). In 2020, Americans legally wagered $21.5 billion on sports gambling at licensed sportsbooks, resulting in $1.5 billion in revenue for the sportsbooks (American Gaming Association, 2021). By 2023, those numbers had grown to $119.8 billion and $10.9 billion, respectively (American Gaming Association, 2024). The increasing prominence of sports betting has led to greater interest across many academic fields, including business strategy.
Both strategy and sports betting offer a risk-reward perspective. Still, the potential payouts and losses with the latter are usually easier to calculate and immediately evident, supporting the aims of our research. Indeed, the levels of knowledge and understanding about potential payoffs and losses in the business arena can vary widely. While some sports gamblers play strictly for entertainment, strategists connect their livelihood to their choices (Strong, Reference Strong2024). Despite those differences, notable overlaps exist. On the one hand, gamblers who gamble for entertainment or office bragging rights may be less concerned about winning, which can lead to more irrational choices when they believe certain games are sure bets.
On the other hand, despite notable blind spots (Zajac & Bazerman, Reference Zajac and Bazerman1991), some strategists also make decisions just as haphazardly (Das & Teng, Reference Das and Teng2001; Nobre, Grable, Silva & Nobre, Reference Nobre, Grable, Silva and Nobre2018). Many executives struggle to evaluate the probability of success and failure given myriad knowns, unknowns, and uncertainties (Karelaia, Reference Karelaia2009). Although gamblers have access to information (e.g., odds, injury lists, weather forecasts), upsets regularly occur in the sports world, which further supports the parallels between making strategic decisions for organizations and selecting winners of sporting events.
Many strategists invoke an institutional perspective, assuming that the best path forward reflects the processes spawned by repeated actions and the normalized meanings of those actions (Berger & Luckmann, Reference Berger and Luckmann1967; Scott, Reference Scott1987). A decision is perceived as optimal simply because it has worked most often in the past. Without the data patterns to follow though, those strategists lead firms to less profitable activity (Haleblian, McNamara, Kolev & Dykes, Reference Haleblian, McNamara, Kolev and Dykes2012). Although deviating from norms could offer first-mover advantages (Cirik & Makadok, Reference Cirik and Makadok2023), decision-makers tend to stick to what appears most safe, following industry norms to minimize the possibility of adverse outcomes. However, following the herd does not yield anything unique. Although differentiation can be profitable (Hirshleifer, Hsu & Li, Reference Hirshleifer, Hsu and Li2018; Roquebert, Phillips & Westfall, Reference Roquebert, Phillips and Westfall1996), the uniqueness inherently lacks a path to follow. In other words, deviating from the norms offered by industry experts invites much riskier possibilities. Consequently, strategists may be aware that such deviations should occur but are unsure of how to manage the associated risk. Hence, they recognize the need for some meaningful deviation but fail to navigate the risk effectively.
These similarities between sports betting and strategic corporate behavior present intriguing learning opportunities for business strategists. Specifically, strategies used in betting contests can inform corporate strategists about necessary trade-offs between seeking opportunities that reflect legitimacy and those that establish uniqueness. To investigate these phenomena, we use data from a betting contest called Ranked Pick’em. These contests constitute a betting pool, where all participants pay an entry fee to enter the contest and become eligible for monetary prizes. In most betting pools, participants predict the outcomes of sporting events, and often must assign a weighted value to their selections. We present two approaches that could be used to guide those decisions.
Background
We begin by providing an overview of the NFL schedule structure, the rules for these Ranked Pick’em contests, and foundational knowledge of gambling terms. The NFL currently consists of 32 teams, and each team plays 17 games in an 18-week season. Each team has one bye week in which they do not play a game. In the betting contests, each participant chooses a winner for the games scheduled for that week and assigns a unique point value to each selection (from 16 down to 1 in weeks with no teams on bye). If the participant’s selected team wins, the participant receives the point value they placed on the selection. Contest winners, for a given week and across the entire season, are determined by the highest scorers in that point system. In each season, the Ranked Pick’em contests featured 170 participants, each paying a $60 entry fee, totaling a $10,200 pot of money for each season. According to the contest rules, one-third of the money from entrance fees is paid out to the best-performing participants every week; the other two-thirds is paid out to the best-performing participants on a season-long basis. Table 1 presents the actual payouts for participants in the years under investigation. If participants had tied scores, payouts were evenly split. For example, if two participants tied for third place in a given week, they would each have received $15, from $30 divided by 2; if two participants tied for second place in a given week, they would each have received $45, from ($60 + $30) divided by 2.
Table 1. Payouts to contest winners
The contests whose results we use were run by a private individual who collected no additional fees for managing them. Figure 1 illustrates an example of results from the 2021 season, with participant names redacted.
Figure 1. Sample of Ranked Pick’em contest results from 2021.
Some NFL teams perform better than others over a given season due to having more talented players, coaching prowess, or even luck. These teams are favored by sportsbooks when facing poorer-performing teams. In point-spread betting, bookmakers determine how many points the favored team is expected to win by and allow bettors to choose a side of the point spread on which to wager. For instance, if Team A is favored over Team B by 10 points, the point spread will be displayed as ‘Team A (−10)’. Therefore, when choosing which team will win the game, participants in Ranked Pick’em contests may be naturally inclined to align with bookmakers and resort to selecting the favored teams. Moreover, the stronger the favorite (i.e., the larger the point spread), the more contest points participants may be inclined to assign to it, given that bookmakers are generally astute at assigning point spreads to games.
Decisional approaches
When data suggest a particular team will win, it makes sense to select that team. When spreads are greater, it makes sense to assign higher confidence and higher contest point values to higher favored teams. This approach appears less risky because it strictly adheres to conventional standards outlined by industry experts (i.e., bookmakers). The safer, risk-averse decisional approach leans completely on the information bookmakers provide. Before electronic displays became common at horse races, when more bets were placed on a horse, their odds would change, and the current odds would be erased and updated. Multiple iterations of that process would leave considerable chalk dust, and it conditioned bettors and bookmakers to refer to odds-based favorites as chalk (Santaromita, Reference Santaromita2022), a reference that persists today. In line with this conventional terminology, we use Chalk to refer to a decisional approach whereby industry norms are used to make decisions. Specifically, if industry data suggest that one team would win and the point spread is large, contest participants would pick the favored team to win and assign it high contest point values.
Because Ranked Pick’em is a zero-sum competition, absolute performance is less critical than relative performance. Given the weekly payouts throughout the season, a participant could earn a positive profit by scoring highly (relative to the field) in just one or a few weeks, even if their season-long score was low. Within the rules of the competition, two-thirds of the entrance fees are allocated to the season-long winners, drawing the most attention from participants. Making safe choices based on industry standards may avoid criticism, but picking the favorites dilutes potential performance (e.g., payouts). Therefore, to explore how decisions might systematically deviate from the norms, we evaluate a second decisional approach we call Contrarian. This approach uses the same formula as Chalk to pick game winners and assign their contest points, with one minor adjustment. Like Chalk, nearly all favored teams are selected, and contest point values are ordered by size of the point spread. However, with Contrarian, the team bookmakers expect to lose by the biggest margin (i.e., the biggest underdog) was selected to win the game instead of the favored team, while maintaining its maximum contest point selection. For instance, if 16 games occurred in one week, 15 of Contrarian’s selections (both in terms of which team wins and how many contest points were assigned to the game) would mirror Chalk’s selections. The only difference would be Contrarian’s choice to flip the selected winner of the game with the most contest points, choosing the underdog instead. Such a deviation blends novelty and legitimacy, as nearly all the norms are followed, but with one crucial and systematic exception. Table 2 presents a representative example of the participants’ selections of the most favored team in a given week in the contests. It uses the data from Week 8 of the 2017 season, where the Philadelphia–San Francisco game had the biggest point spread of the week, with Philadelphia as the favorite to win. Only one participant picked San Francisco to win the game, and for only four contest points. Every other participant picked Philadelphia to win, and nearly 83% of the participants (141 out of 170) assigned the selection of Philadelphia either 16 or 15 contest points. In moving from Chalk to Contrarian, a participant would be flipping the selection from Philadelphia for 16 points (something more than 100 other participants did) to San Francisco for 16 points (a unique choice). That novelty is the defining characteristic of the Contrarian strategy.
Table 2. Frequency and contest point value of biggest point-spread favorites being selected in an example contest week (week 8 of the 2017 season)
Biggest Point Spread: PHI (Philadelphia) over SF (San Franciso).
From a business perspective, Chalk is rational and ostensibly risk-averse because it follows market signals perfectly. However, assuming a general tendency for competitors to act similarly, participants pursuing a Chalk strategy minimize uniqueness and most likely wind up in a crowded field with modest outcomes and limited, if any, payouts. In a business setting, Chalk-like, ‘rational’ firms minimize differentiation, a key driver of firm performance (Banker, Mashruwala & Tripathy, Reference Banker, Mashruwala and Tripathy2014; Gong, Yu & Huang, Reference Gong, Yu and Huang2021; Stonehouse & Snowdon, Reference Stonehouse and Snowdon2007). To illustrate how and where firms might depart from the norms, Contrarian adopts primarily market-driven choices, deviating only slightly. However, those deviations promote uniqueness by targeting areas that other participants would least expect, as they select underdogs in the highest point spread games and assign them the highest point values.
We revisit historic NFL data (i.e., point spreads and game results) and the results of real Ranked Pick’em contests to assess what monetary outcomes would have been had our two decisional approaches been used consistently. Exploring these results holds promise for strategists, as it may illuminate how and where systematic deviations from industry norms can mitigate risk and increase potential performance.
Data
We applied our two approaches to data from all the weekly and season-long Ranked Pick’em contest results from the 2014–2022 NFL seasons. We defaulted to picking the home team to account for instances where the point spreads were zero. Additionally, if multiple games in the same week had the same point spreads, we randomly broke the ties regarding the value we assigned to each game. To limit the randomness of assigning point values to games with equal point spreads, we simulated the possible outcomes of participant scores, given the varied ways to split ties, with 10,000 iterations each week of each season to generate an exact empirical distribution of scores that the rationale could have had, depending on how ties were broken. It is essential to note that this analysis did not simulate any aspects of the Ranked Pick’em contest results or NFL game results; instead, it utilized the actual outcomes of these events. The only ‘simulations’ involve the way that games with equal point spreads were randomly assigned contest points in our decision approaches. In other words, the results of our approaches in any given week are entirely deterministic, except for the randomness in breaking some ties that the strategies themselves do not address. The simulations of how different selections fared were run separately by week within each season.
Table 3 illustrates one possible selection based on the Chalk rationale for the Week 18 games of the 2023 season. Games are listed by the size of the point spread. Equal point spreads occurred at 4, 3.5, and 3 points. Therefore, those tied spreads were broken randomly in the way those games were ranked for contest point purposes. Selected teams are shown in the second column, and bolded if the team won the game. In this example, despite many games having the same spreads, there are only two possible outcomes in terms of the total score for the Chalk rationale, and both were equally likely. The selection shown would have scored 89 points that week. Both favorites in the 4-point spread games lost, so it would not matter how the tie was broken when assigning contest point values to each selection. Likewise, every favorite in the 3-point spread games won, so it would not make any difference how the tie was broken when assigning values to each of those games. However, different game outcomes occurred for the 3.5-point spread games concerning the favorite. Detroit won as a favorite, but Los Angeles lost as a favorite. The selection shown in Table 3 randomly assigned Detroit the higher point value over Los Angeles (10 instead of 9). If the point assignments for those two games were reversed, as would happen half the time in our simulations, the selection would have scored one fewer point (88 instead of 89). The Contrarian rationale for the same week would have resulted in either 72 or 73 points. It would have selected Washington instead of Dallas for the highest value (16 points), and since Dallas won, it would have performed 16 points worse than Chalk in every simulation. Note that regarding Contrarian, multiple games with the same biggest point spread in a given week are rare. More importantly, there was never a situation (in any week in any of the nine seasons investigated) where such tied point spreads had differing win/loss outcomes concerning the favorites in those games.
Table 3. One possible selection of chalk for week 18 of the 2023 NFL season
Total points earned for this selection is 89.
Bolded teams won their game.
We evaluate the earnings of our two strategies as follows. First, for a given week for a given strategy, we compute the following summary statistics of the empirical distribution of the simulation results: maximum, third quartile (Q3), median, first quartile (Q1), and minimum. Because we are evaluating the historical performance of these deterministic strategies against real contest outcomes, the use of inferential statistical approaches at this point (such as constructing a confidence interval for the score in a given week) is not justified. Instead, the summary statistics can represent the amount of ‘luck’ a participant would have had when breaking the ties of equal-point-spread games when assigning contest point values. For example, the maximum score would only have been realized if the higher point values were chosen for the winning teams every time equivalent point spreads saw different results in terms of favored teams winning or losing their games. In other words, the participant’s maximum score required maximum luck in how those contest points were randomly assigned. We then check how much money those summary statistic scores would have earned for that week using the actual results from other participants from the Ranked Pick’em contests. For the season-long results, we use the first through final simulations in each week of the given season, sum those scores for each numbered simulation across the weeks, and then use the same summary statistics as the results.
Results
Since all results are presented as profits, $540 has been subtracted from the sum of any payouts earned ($60 per season entry fee for nine seasons). Somewhat surprisingly, no single simulation of Chalk ever won a weekly payout. Notably, all of its payouts are based solely on season-long results. On the other hand, as expected, Contrarian was never competitive for the season-long payouts. All the payouts it would have earned came from weekly results (the same summary statistics are shown for the results of this rationale, though they are computed on a weekly basis). Table 4 shows the profits each rationale would have realized across the nine seasons of competition, depending on the luck involved in breaking ties when assigning points to selected winners.
Table 4. Profits of tested approaches over nine seasons
The Chalk rationale had one particularly good season (2017). In that year, its maximum score would have won first place for the season-long payout, and its Q3, median, and Q1 scores would all have come in fifth place. In only one other season would any summary statistic other than the maximum have resulted in a payout (a Q3 finish in ninth place in 2019, which helped Q3 realize a positive profit of just $107 across the nine seasons). The maximum Chalk score would have received a payout in seven of the nine seasons (all but 2015 and 2016). Still, that summary statistic assumes a very fortunate outcome among the many possibilities.
Most importantly, median Chalk would have lost $64 across the nine seasons of this contest. Q1 would likewise have lost $64, and minimum Chalk would have lost all entry fees ($540), never once receiving a payout. To summarize, the range of possible outcomes for the Chalk rationale is relatively high. It could have lost as much as $540 or won as much as $3,577.50 over the 9 years of competition, and its median outcome would have been a slight loss of $64. Its difficulty in generating positive profit at anything below the maximum outcomes stemmed from being too similar to other participants’ selections.
In contrast to Chalk, the Contrarian results show a much lower upper bound for profit but, importantly, higher earnings across the rest of the distribution. Even its minimum possible scores would have resulted in a positive profit of $120 (or 22.2% of the entry fees paid) across the nine seasons. Compared to Chalk, Contrarian had a higher minimum ($120 vs. −$540), Q1 ($165 vs. −$64), median ($195 vs. −$64), and Q3 ($323.33 vs. $107) profit. Chalk only dominated Contrarian at the maximum profit ($3,577.50 vs. $475). In sum, the seemingly ‘riskier’ strategy (Contrarian) generated small but reliable long-term positive profits. In contrast, the more balanced Chalk strategy offers a chance at a huge payout, largely due to the results achieved in just one season. Because that one season could very well be non-replicable, Chalk was more likely to lose money than gain money over the years of the competition.
Discussion
Just as firms develop different strategies to pursue myriad missions and goals (Foss & Lindenberg, Reference Foss and Lindenberg2013), each rationale was aligned to consider varied goals and optimal outcomes. Chalk was a rational approach to decision-making. More specifically, it reflected a low-risk perspective that resembles the nature of decision-making at many businesses, where the goal is to survive, not necessarily thrive or become industry leaders. Such may also be the case for certain gambling contests, such as March Madness office pools, where some participants care less about monetary payouts but want to avoid making outlandish choices that invite criticism from coworkers. Because it is relatively safe, Chalk would be expected to perform decently well over an entire season, just not well enough to generate high returns. Contrary to this thinking, Chalk never placed for payouts in any week during any of the nine seasons, and it was only truly competitive for a season-long payout once. Thus, following industry rules could have harmed overall performance to the point of incurring losses (except in Q3 and above), and the rationale only rarely yielded outstanding results.
On the other hand, Contrarian afforded no hope of competing for the season-long prizes. Its highest-valued selection would often lose the game and earn zero points, as expected, at least according to industry standards (i.e., betting odds). Thus, neglecting industry-standard ideals would be viewed as riskier than the perspective afforded by Chalk. This is one possible reflection of the entrepreneurial desire to deviate from the industry rules. More specifically, it demonstrates how industry standards are often followed yet still allow for the pursuit of uniqueness, despite apparent risks. Contrarian would fare quite well (i.e., positive profit per season) with only one or a few wins of the weekly payouts. Moreover, it could fail to win a payout nearly every week, yet it will still be successful in the long run. In short, payouts in weeks won would outweigh the statistical likelihood of the earnings possible from the season-long earnings afforded by the Chalk rationale. Indeed, we demonstrate that the seemingly riskier Contrarian strategy yields a higher floor (minimum), higher Q1, higher ‘expectation’ (median), and higher Q3 of long-run profits in Ranked Pick’em than Chalk, albeit with an admitted lower ceiling (maximum). At the same time, strategists might assume that pursuing uniqueness at the expense of legitimacy only offers rare but large payouts; nearly a decade of actual outcomes presented here suggests otherwise.
Applications for business strategy
Our results illustrate how and where decision processes can blend legitimacy and novelty. They show how pursuing conventional wisdom grounds most decisions in industry norms. They also show where systematic and intentional deviation from those norms may significantly increase performance opportunities. Strategic decision-makers must strike a balance between uniqueness and legitimacy. Indeed, there are benefits and costs associated with pursuing an ostensibly risky strategic course of action. Conversely, if a firm’s goals extend beyond survival, continued pursuit of actions that merely mirror the best practices of the industry may not yield the desired outcomes, as was the case with Chalk. Despite the safer nature of the choices, profit potential was lower because many competitors often followed the same approach. Using an analogy between sports betting and competitive business activity, we compare gamblers and payouts in a unique competition to firm strategies and business performance. Findings from our two decision approaches mirror a paradox in strategic decision-making, where managers seek to maximize desired outcomes but also limit risk. That risk, however, affords the space to differentiate business offerings from competitors. At the decision level, managers are trained to identify and select actions that maximize possible outcomes and adjust for risk. At the strategic level, executives seek to craft and execute an approach that maximizes the organization’s outcome based on its mission and goals. Markets often reward differentiation, which is accomplished when a strategy – a collection of decisions – deviates from the norm in some way. Therefore, strategists must determine which industry rules to follow and which to break. Our results offer some clarity and justification for making those decisions.
Several critical applications exist for strategists. First, managers cannot wholly rely on rational options if they want firms to do more than survive. Just as Contrarian did not always select favorites, managers should not automatically accept industry-generated best practices. Consistently doing so likely fails to develop unique combinations of resources and capabilities, resulting in poor differentiation, crowded industry domains, and poor performance compared to industry leaders. To achieve distinctiveness and enhance performance, managers must make some decisions that deviate from conventional wisdom. Although deviations should follow a clear reasoning rather than relying solely on intuition (Cristofaro, Giardino, Camilli & Hristov, Reference Cristofaro, Giardino, Camilli and Hristov2024; Elbanna & Fadol, Reference Elbanna and Fadol2016), guidance on how and when those deviations should occur remains less understood. Contrarian showcases how minor deviations from the norm can significantly influence the novelty of strategies.
Consciously betting against the odds seems counterintuitive, and selecting several risky opportunities at once is likely to hinder strategists. Contrarian shows just one way to deviate from the norms, but market instability invites managers to distinguish their businesses in several ways, opening doors for improved performance (Qian, Yang & Li, Reference Qian, Yang and Li2016; Wang, Dou, Zhu & Zhou, Reference Wang, Dou, Zhu and Zhou2015). With so many options, managers still struggle to select where their novel decisions might be best employed, which happened when Contrarian made selections for the highest underdog. Indeed, market turbulence heightens strategic uncertainty, which creates opportunities (Chen, Wang, Huang & Shen, Reference Chen, Wang, Huang and Shen2016; Leppänen et al., Reference Leppänen, George and Alexy2023; Wilden & Gudergan, Reference Wilden and Gudergan2015).
Recall the details shown in Table 2, where the biggest underdog in a given week was selected to win by less than 1% of the field (i.e., only one participant, and not for a large point value) even though the biggest underdog of the week won 28 times over those nine seasons, or nearly 19% of the time. Strategists looking for largely overlooked opportunities could do much worse than systematically investing in something that ‘hits’ 19% of the time but is considered by only 1% of the competition. For example, firms can leverage novelty by developing new products, integrating supply chains, or entering new markets (Xu, Li, Sun & Zhao, Reference Xu, Li, Sun and Zhao2012). Similarly, since turbulent markets encourage innovation, a precursor to competitive advantage in many firms (Chen et al., Reference Chen, Wang, Huang and Shen2016; Rodrigo-Alarcón, García-Villaverde, Parra-Requena & Ruiz-Ortega, Reference Rodrigo-Alarcón, García-Villaverde, Parra-Requena and Ruiz-Ortega2017), more opportunities exist in areas that few people tend to explore. This helps explain why small, innovative firms often perform well in turbulent environments (Kraus, Rigtering, Hughes & Hosman, Reference Kraus, Rigtering, Hughes and Hosman2012; Spriggs, Yu, Deeds & Sorenson, Reference Spriggs, Yu, Deeds and Sorenson2013).
Contrarian and Chalk also inform our understanding of strategic agility. Agile organizations may efficiently and effectively redeploy/redirect their resources to create, protect, and capture value (Khan, Majid & Yasir, Reference Khan, Majid and Yasir2021; Margherita, Sharifi & Caforio, Reference Margherita, Sharifi and Caforio2021; Srinivasan, Srivastava & Iyer, Reference Srinivasan, Srivastava and Iyer2020). Agility is costly because it commits resources to develop and maintain dexterity. Some firms succeed by minimizing agility and costs, while others leverage agility to promote innovation and crisis response (Fan & Xiao, Reference Fan and Xiao2023; Madadian & Van den Broeke, Reference Madadian and Van den Broeke2023; Parnell, Reference Parnell2021; Teece et al., Reference Teece, Peteraf and Leih2016). Managers must know when agility is suitable for their organization and how much is necessary. The NFL’s weekly schedule consists of one game of the week played on Thursday (before the remainder are played on Sunday and Monday). The Ranked Pick’em contest permits participants to change their selections and point values until the respective games begin. If Contrarian correctly selects the week’s biggest underdog as the winner of the Thursday game, the rest of the week’s picks could be made more agile, as the selection would put the participant far ahead of the others.
While successful strategies often emerge from heuristics embedded in rational decision-making, they may not always be tightly coupled to decisions. While Chalk realized small payouts, most of the underlying selections made were correct. Some NFL teams outperformed others due to having better players, coaching, or simply good fortune. Similarly, some businesses outperform their competitors due to superior talent (i.e., human resources), effective strategies, and luck (Parnell & Dent, Reference Parnell and Dent2009; Parnell, Dent, O’’Regan & Hughes, Reference Parnell, Dent, O’Regan and Hughes2012). Sometimes good fortune cannot be explained rationally, and bad fortune cannot be avoided.
Finally, strategic success depends on firm goals, metrics, and orientation. For example, some gamblers might be content with performing near the middle of the pack in the short run, while others view gambling as a form of investing and are focused on net gains over the long term. Similarly, a small family business might be content functioning at a modest profit and serving the community over the long haul, while a startup might be willing to operate in a turbulent, risky environment in search of a big payoff (Alonso, O’Brien, Kok & Kok, Reference Alonso, O’Brien, Kok and Kok2018; Anderson, Bergbrant, Hunter & Reeb, Reference Anderson, Bergbrant, Hunter and Reeb2023; Strakova, Reference Strakova2024). Firms can benefit from considering their strategies as a series of decisions, where success is evaluated as the sum of its parts, not whether each decision was a win or a loss. Someone in a Ranked Pick’em who only seeks to avoid embarrassment (i.e., a form of legitimacy) in an office pool might pursue Chalk over Contrarian despite reduced profit potential. Our data support the complex nature of strategic choice, considering the varied goals of decision-makers.
Conclusions, limitations, and future research
We applied two reasonable approaches to strategic decision-making in NFL Ranked Pick’em contests over nine seasons, based on actual outcomes. Chalk, the risk-averse strategy, is akin to legitimacy. Contrarian, the ostensibly riskier alternative, leverages uniqueness and differentiation. Both strategies demonstrated profit potential during that time, but Contrarian yielded more reliable and consistent overall profits than Chalk. These findings from a real betting contest suggest that a string of entirely rational decisions does not necessarily maximize performance, a principle that also applies to strategic decision-making. Our analysis demonstrates that risk-averse decisions, when evaluated collectively, rarely offer enhanced performance potential, even when each individual decision is technically rational and correct. Industry-based best practices can guide decision-making and strategy creation, but careful deviations from the norm can help managers make more effective decisions.
These wagers demonstrated that heuristics grounded in, but not necessarily married to, perfect rationality can produce higher long-term performance. Notably, industry best practices are helpful starting points, but strategists should learn to deviate from conventional wisdom in ways that create value for their firms. While many strategists recognize this reality, assessing how and where such deviations should occur is often difficult. Our data highlight those spaces, showing that even a minor deviation, when few competitors are likely to select the same deviation, can create a significant upside in the long run. Choosing those options helps facilitate strategic agility, transforming a series of mostly rational decisions into a coherent and distinctive competitive strategy. Several limitations to this study should be recognized. First, we evaluated the profitability of these strategies ex post and outside of the contest venue. Had we been making decisions guided by either the Chalk or Contrarian rationales over a more extended period in the actual contests, other participants may have identified and mimicked our strategies. If such imitation occurred, in the case of Contrarian, profit potential would be significantly reduced with even one or two other participants following the same strategy. Indeed, mimicking successful strategies, even seemingly odd ones, occurs regularly in business competition. Furthermore, the limitation suggests the need to continue searching for options not recognized by competitors if advantages are to be sustained.
Second, sports betting is often less than a zero-sum game for gamblers due to the vigorish, or juice, charged by sportsbooks. That is, most participants, despite winning sporadically, lose money in the long run. While performance variation within industries is common in business, total industry profit – as defined by analysts – is often positive, but only occasionally negative. In market competition, it is possible and common for most competitors in an industry to achieve their goals and perform well. Thus, varied goals and zero-sum natures limit direct comparisons of our data to market competition. Finally, the case analysis presented herein evaluates only two decision approaches and focuses on financial performance as the primary outcome. Indeed, endless strategies are possible in both business and gambling arenas. In addition, many firms view business performance from a balanced scorecard perspective, where various financial and nonfinancial goals are weighted differently from other firms.
Future research should consider how success and failure with other approaches to sports betting – not just Chalk and Contrarian in this Ranked Pick’em contest – can be applied to strategic decision-making. Although business strategy is not a zero-sum game, it bears several important similarities to sports betting. In both instances, decision-makers consider trade-offs, evaluate risk, and invest real resources. Sports betters and strategic decision-makers seek optimal returns, broadly defined (i.e., not necessarily the highest expected financial payoff). For strategic managers, trade-offs often exist between novelty and legitimacy. An evidence-based framework like ours would help them understand and evaluate trade-offs, leading to more effective decisions. Since our framework focused solely on identifying how and when deviations from the norm could prove helpful, decisional approaches to managing trade-offs were not employed. We believe scholars should find new ways to apply lessons from sports betting to business strategy where applicable.
For example, some gambling competitions allow participants to have multiple entries. Researchers might build and deploy rationales reflective of portfolio approaches in those arenas, which could be depicted as a portfolio rationale. Several studies (e.g., Bergman, Cardonha, Imbrogno & Lozano, Reference Bergman, Cardonha, Imbrogno and Lozano2023; Decary, Bergman, Cardonha, Imbrogno & Lodi, Reference Decary, Bergman, Cardonha, Imbrogno and Lodi2024) have begun investigating strategies for competition that afford those approaches, and future investigations would be more novel if they adopted the portfolio rationale in conjunction with our Contrarian-like rationale, considering the diversification of risk. Outside gambling arenas, future investigations could consider patent filings from businesses in the same strategic group. If several companies pursue patents in the same industry, a Contrarian-like rationale would suggest uniqueness may exist through unrelated patents. In other words, if most businesses emphasize similar innovative efforts because industry data suggest their importance, continued commitment to that area reflects a Chalk-like rationale. More profit potential may exist by selecting unrelated industries in which to deploy innovative resources, as doing so could reflect both Contrarian-like and portfolio-like rationales.
In addition, other strategies within the Ranked Pick’em contests could be evaluated vis-à-vis business strategies. One example is Contrarian-like strategies that select underdogs based on the expectation of other participants’ picks and the point spread, rather than the rank or size of the point spread in a given week. Another depends on the aforementioned quirk of the Thursday game on the NFL schedule, which might allow for a strategy that changes its risk profile depending on the result of the Thursday night game. Suppose a participant chose the underdog team in the Thursday game for 16 points (a somewhat risky strategy, and different from Contrarian unless that was the week’s biggest point spread game), and that team wins. In that case, the participant might then shift to the less risky Chalk-like strategy for the remaining games that week. If their Thursday selection wins, they get out ahead early and then relax to avoid taking too much risk. Conversely, if that Thursday night underdog loses, the participant may have to pursue even riskier selections (heavily over-picking and over-valuing underdogs) for the remaining games. Those actions, specifically in the switching context, could spark discussions about the advantages and disadvantages (Kopel & Löffler, Reference Kopel and Löffler2008; Makadok, Reference Makadok1998).
Similarly, participant strategies could change throughout the season. Indeed, Contrarian has no chance of being competitive for the season-long prize due to its highest-valued selection losing in most weeks. Suppose a participant played Contrarian for the first week of the season, the largest point spread game ended in an upset, and the underdog won, and the participant won the first-place weekly prize. Should the individual then switch to the more risk-averse rationale (i.e., Chalk) for the remainder of the season, knowing they are ahead of the other participants? On the other hand, participants who are out of the running for the season-long prizes toward the end can and should switch to riskier strategies (such as highly ranked upsets) to win a weekly prize. While our findings showcase appropriate ways to locate when and how to deviate from industry expectations, the results also suggest several intriguing avenues for further investigation. We invite others to extend these efforts and suggest that management scholars develop other empirically based decision-making approaches that might be tested against sports betting data.
Open access
